Report No. 36483-NG Nigeria Competitiveness and Growth Country Economic Memorandum (In Three Volumes) Volume III: Technical and Statistical Annexes May 30, 2007 Poverty Reduction and Economic Management 3 Country Department 12 Africa Region UK DFID Document of the World Bank TECHNICAL ANNEX 1-A: GOVERNMENT'SECONOMIC REFORMS Over the last few years Nigeria has made considerable progress in stabilizing the economy, implementing structural reforms, and strengthening mechanisms and institutions for accountability. Key macroeconomic reforms include the introduction in 2004 of an oil price based fiscal rule. This sterilizes oil revenue proceeds above a benchmark price o f oil set to reflect historicaltrend in oil prices. All extra revenue derived from the difference between actual and predetermined prices are held in a special "excess crude oil revenue" account at the CBN. The application of this rule has resulted in considerable budget surpluses and quick accumulation o f foreign currency reserves. At the end o f 2005, the gross internationalreserves amounted to US$28 billion, about 4 times above the level of reserves at the end of 2003. This rule has greatly improved the conduct of fiscal policy and i s beginning to break the historical strong link between oil revenues and the stance o f fiscal policy. Improved fiscal performance allowed the Government to reduce its reliance on the Central Bank of Nigeria (CBN) for deficit financing with a sizeable decline in CBN lendingto the Government registered between 2003 and 2004.This has been an important factor inkeeping inflation broadly under control, despite a massive inflow o f foreign exchange. Although in 2004 macroeconomic stability was achieved primarily through fiscal policy means, tighter monetarv policy in 2005 contributed to maintaining on average a lower inflation environment. The Government and CBN were also successful in completing a major consolidation o f the banking sector in late 2005, triggered by raising the minimum capital requirements for commercial banks from about US$15 million to US$190 million. Through mergers and acquisitions, 25 much larger banks have emerged out of 89 original banks. This consolidation i s expected to facilitate deepening of the domestic market for government debt. A new fully-fundedpension system was launched in July 2005 replacing the unaffordable pay-as- you-go system for public employees. The introduction of this new system aims at stopping the country's chronic problems with pension arrears and is expected to facilitate gradual improvements inthe inefficiency of social security and capital markets. Progress towards the unification of the exchange rate market has been substantial over the past three years with the Government and CBN taking steps to increase both exchange rate and interest rate flexibility, including the successful implementationo f a wholesale market for foreign exchange inFebruary 2006 and further liberalization ofthe exchange rate market inMarch2006. Structural measures include public expenditure management reforms related to budget formulation, execution and reporting, measures to improve transparency and management o f public resources and their utilization, civil service reforms, privatization (particularly that of utilities), trade reforms (including custom and port reforms) and energy sector reforms. Progress in these areas has been positive over the past three years and reforms with high-levelpolitical commitment been renewed regularly. Progresshas beenparticularly encouraging inareas such as transparency in the distribution of oil revenues to different tiers of government and in the financial flows in the oil sector. Other areas of progress include procurement reform, and concessioning o f the ports. Inearly 2005, momentum also picked up inareas that had previously recorded relatively slower progress, like trade policy (with the adoption o f the ECOWAS Common External Tariff), privatization and civil service reforms. Legal backing to some important reforms in key areas has either been cleared (e.g., new tariff structure) or is pending 1 approval (e.g., fiscal responsibility, public procurement, extractive industries transparency and tax administrationreforms). As part of budget reforms the FederalMinistry of Finance institutedthe preparationof an annual Fiscal Strategy Paper outlining the fiscal assumptions, targets and spending priorities for the fiscal year to guide government expenditurepolicy. The FederalGovernmenthas also introduced an MTEFprocessand severalministriesare preparingcosted mediumterm strategies in line with NEEDS (Nigeria's PRSP) to form the basis of their budget proposals. Further, a virtualpoverty fund is being set up to track poverty-reducingspending. A new budget classificationsystem - more disaggregatedfor capital spending -was also introducedin2005 but still needsrefinement to reach internationalstandards. Disbursements against the budget are authorized by a newly established high level Cash Management Committee and some limited improvements have also been made in the accounting area through early introduction of pilot small scale transaction recording and accounting systems (TRS) in a number of MDAs, the recruitment of additional 2004 to instituteprocurement reforms in the use of the federal capital budget - covered over 80 staff and training. The Budget Monitoringand PriceIntelligenceUnit (BMPIU) - establishedin percent of the 2004 federal capital budget and its efforts are being sustained with the aim to reduce by 50 percent the turnaround time for reviewing procurement documents. The Public Procurement Bill, which foresees the establishment of an independent Public Procurement Bureau, was passed by the House or Representativesand is awaiting passage by the Senate. To strengthen the management of public accounts, the Government plans to computerizethe Budget Office and the Office of the Accountant General of the Federation and interconnect their operations with those of the BMPIU and CBN. The backlogs of financial reportinghave been substantially reduced with Auditedreports beingproduced up to 2003 (partial). The government made also some improvements in its reporting of progress with the execution of the annual budget but bothperiodic and annual reportingfor budgetmonitoringand accountabilitypurposes remainsweak. On the revenue front Nigeria leads worldwide on the implementationof the ExtractiveIndustries Transparency Initiative (EITI) and following the release of the audit reports of the oil and gas sector for 1999-2004last April 2006, the Government is planningto sustain the audit effort and cover the year 2005 as well as issue quarterly audits for 2006. Further,the publicationof revenue allocations fromthe FederationAccount to the three tiers of Governmentbegan in 2004 continues and now covers the period from 1999 to date. Major efforts are also being undertaken by the Federal InlandRevenue Service (FIRS) to reformthe revenue administration, through a process of restructuring, modernization and automation which is progressing, albeit slowly. The FIRS succeeded in re-organizingitself according to functions and is now streamliningactivities in all its decentralized offices. The FIRS is undertaking a nationwidetaxpayer enumeration scheduled to be completed by the end of 2006 to prepare the ground for an integrated tax administration system which would include Individual Tax Identification Number (TIN). Several bills are before the legislature to back the future semi autonomous status of the FIRS, hence a repositioningof the Agency in line with internationalgood practice, and to introduce tax policy and administrationamendments. The Government embarked in 2004 in a process of civil service reforms in 6 pilot ministries which are being broadened to cover the rest of the civil service; guidelines for structuring of ministries and parastatals have been adopted by the Federal Executive Council for the restructuringof an additional5 key ministriesintargetedby end 2006, bringingthe total number of ministries being reformed to 11. An IntegratedPersonnel and Payroll Information System (PPIs) is being introduced to help monitoring staffing numbers in the civil service and deployment ofthe first phaseofthis beguninlate2006. 2 Despite a relatively slow progress the privatization effort has been revamped with the change in leadership of the Bureau of Public Enterprises (BPE) in March 2005. A number of enterprises have now reached the point of sale and the legal ground for the privatization of utilities i s being prepared. Ten enterprises are listed for privatization or concessioning in 2006 and early 2007 while the establishment o f regulatory authorities for the relevant sectors to provide oversight functions for these sectors is continuing apace. Six brick and clay companies will be offered for sale by end April 2006. While 51% of Hilton Hotel was sold in 2005, the P O for Government's 49% will take place by the end o f 2006 just like the offer for sale o f L e Meridien-Sofitel Hotels. NITEL is being repackaged for a new offer following the below-reserve price bid received in December 2005. Particularly important giventhe country's chronic power and fuel shortages has beenthe progress on reforms in the energ sector. In March 2005, the Electric Power Sector Reform Bill was signed into law, providing the legal underpinning for the unbundling o f the Power Holding Company of Nigeria (PHCN) into 18 successor companies and the setting up of a regulatory commission, the NationalElectricity Reform Commission (NERC). Arrangements are now under way to privatize 3 o f the companies by the first quarter of 2007. While the BPE is leading the reform and privatization transaction effort, NEPA and the Ministry of Power and Steel are working to deliver quick-wins and better sector performance in the shorter-term to complement the longer-term reform plans, through better power supply, customer service and financial improvements. Further, with the aim to increaseNigeria's oil refining capacity, the Port Harcourt refinery will be offered for sale with bids opening expected by June 2006 and 11 oil service companies will also be offered for sale by the Autumn of the same year. The Government also succeeded in taking forward the deregulation of the downstream oil market. Most fuel subsidies have been phased out although some price support still remains because o f the high level of international oil prices. The Government has taken steps in the 2006 budget to explicitly reflect the price support inthe budget with all tiers of government contributing to the petroleum support fund ofN150bn. The Government is committed to reforming its trade uolicv regime and to harmonize Nigeria's tariff structure with ECOWAS by January 2008. Following the adoption of the five-band customs tariff schedule inOctober 2005 under the ECOWAS Common External Tariff (CET), the Government will work towards reducing or eliminating the list of banned items by January 2007 or such date as i s consistent with the date for ECOWAS convergence criteria. The 50% tariff band will be reviewedby the end of 2007 following a study on the impact o f the tariff reform on the economy. Further, substantial progress has been made in the areas of custom and port reforms. Following the dismissal of senior custom officials with corruption charges and the establishment of a task force driving custom reforms (modernization o f custom procedures and updating of custom data), the reform of the Nigerian Customs Service is expected to see a sharp acceleration; the fast-track window already created for large importers/exporters will be expanded to handle 40% o f the trade (by value) by end 2006, with a view to reducing by half the average number of days it takes to clear goods from the ports. Onthe other hand, following the successful concessioning o f a number o f ports in 2005, the Nigerian Ports Authority (NPA) is being restructured into two autonomous agencies, which i s expected to translate into improved supervision, management and facilitation of ports operations. The central railways corporation will be concessioned inthe Zndquarter of 2006 followed inthe 3`d quarter by the concessioning of the Abuja Airport. 3 TECHNICALANNEX 1-B:FACTORPRODUCTIVITY ANALYSIS Nigeria has suffered from considerable economic volatility. From the mid-197OYs, after the OPEC-related oil price shock, Nigeria even appears to have experienced a significant amount of technological regress. Intheory, ifHarberger's (1998) logic is extended to cases inwhich there is absence of technological progress, this would suggest that new factors o f production are not emerging, and that duringperiods oftechnological regress, more specialized factors of production may be disappearing. As a crude measure of this it is important to note that the share of services as a percent o f GDP has been declining since 1970, just as petroleum was becoming more important. To calculate total factor productivity, a variant of the method adopted more recently by Bosworth and Collins (2003) i s used. The basic variables neededto undertake this task are GDP inreal local currency units per worker, or, inwhich the labor force is defined as the population aged 15 to 64 years old. While Bosworth and Collins estimate the capital stock in their measure o f capital per worker, or k, = K,/L, , using the perpetual inventory method, when it is applied to Nigeria, total factor productivity exhibits much more regress, due largely to the first period value o f the capital stock. Instead of usingthis methodology, for the initial period the capital stock i s assumedto be one and a halftimes the value of initial GDP (GDP in 1960), and for comparison, a capital-output ratios of 2 i s also shown for Nigeria. There are some data issues to note before moving on. While it is more common to use real local currency gross fixed capital formation (GFCF) to measure investment, data was unavailable for Nigeria before 1973, and for other countries it is not available until later. So, gross capital formation (GCF) is used instead for all countries to get a longer view. Also, real local currency series are incomplete for Nigeria (observations for 1999 and 2000 are missing), but the nominal series is complete, so the GDP deflator is usedto generate the real investment(GCF) for all countries covered here. Finally, Bosworth and Collins (2003) also include human capital intheir calculations, which they estimate by multiplying the country's labor force by a rate o f return to education raised to the exponential power of the average years of schooling. Assuming a rate o f return o f 7 percent to education across all countries, the human capital stock is computed as: H, =L, (1.07)avg.yearsof schooling * Human capital per worker simply computed by dividing this term by the labor force, or h, = H,/L, = (1.07)avg.years of schooling, which is simply the assumed rate of return to human capital. The key to estimating human capital for each country is to get an estimate for the average years of schooling. Bosworth and Collins use the Barro-Lee data for years o f schooling.', Since 'Since data are only available at five year intervals, starting in 1960, and ending in2000, a linear interpolation method i s usedto fill in points for observations associatedwith years in which there is no data for each country. This method resultedinthe first value for Beninbeing negative, so that initial value is replacedinstead with a value of one half of the value for 1961,the secondyear inthe sample for Benin, or 0.001. The Barro-Lee dataset is available from httr,://www.cid.harvard.edu/ciddatdciddata.html. However, since there is no data for Nigeria, the Nigerian average years of schooling is computed as the average of the average years of schooling reportedfor Benin, Cameroon, Ghana, Niger, and Togo. This seems reasonable since the sparse data on enrollment rates available from htto://wWW.uis.unesco.orgiurofiles/EN/EDU/countn~Profile en.asux?code=5660 suggest that there is little difference ineducationinNigeria and neighboring countries inthe region. 4 this data is available at yearly intervals, empty observations can be filled by a linear interpolation of the trend. With these basic variables, it i s necessary to compute the growth rates o f output per worker, capital per worker and human capital per worker. From these growth rates, total factor productivity (TFP) i s computed simply as follows: = _-0.35 .--Ak 0.65 A h e- Yr k , hr since it is assumedthat physical capital's share o f income is 35% while human capital's share of income is 65 percent. Incomputing these growth rates the initial observation for each o f the three series is lost. Table 3 details the contribution o f each factor of productionto GDP. What is telling is that over the decades Indonesia had both increases in capital per worker as well as increases in TFP, while Nigeriahad declines in TFP, and negligible increases in capital per worker. In fact, it even looks as ifthere were three "lost decades'' Table 1: Contributions to Total Factor Productivity in Nigeria, 1961-2004 Depreciation=S% K(O)=l.S*GDP(O) K(O)=2*GDP(O) Nigeria Output Per Human TFP Capital Per TFP Capital Worker Capital per Worker Per Worker Worker 1961-1970 2.8% 0.1% 2.2% 0.5% 2.7% -0.1% 1971-1980 2.1% 0.2% -0.5% 2.4% -0.1% 2.0% 1981-1990 -1.8% 0.3% -1.4% -0.7% -1.3% -0.7% 1991-2000 -0.5% 0.2% -0.9% 0.2% -0.9% 0.2% 2000-2004 2.2% 0.2% 1.4% 0.6% 1.4% 0.6% Indonesia 1961-1970 2.2% 0.5% 1.4% 0.3% 1.9% -0.2% 1971-1980 5.2% 0.4% 1.9% 2.9% 2.4% 2.5% 1981-1990 3.6% 0.1% 1.1% 2.4% 1.2% 2.3% 1991-2000 2.2% 0.6% 0.2% 1.4% 0.2% 1.4% 2000-2004 2.9% 0.6% 2.0% 0.4% 2.0% 0.3% Depreciation=l 0% K(O)=l.S*GDP(O) K(O)=2*GDP(O) Output Per Human TFP Capital Per TFP Capital Worker Capital per Worker Per Nigeria Worker Worker 1961-1970 2.8% 0.1% 3.3% -0.6% 3.9% -1.3% 1971-1980 2.1% 0.2% -0.5% 2.4% -0.2% 2.1% 1981-1990 -1.8% 0.3% -0.8% -1.3% -0.8% -1.3% 1991-2000 -0.5% 0.2% -1.0% 0.3% -1.0% 0.3% 2000-2004 2.2% 0.2% 1.3% 0.7% 1.3% 0.7% Indonesia 1961-1970 2.2% 0.5% 2.5% -0.8% 3.1% -1.4% 1971-1980 5.2% 0.4% 1.8% 3.1% 2.2% 2.7% 1981-1990 3.6% 0.1% 1.2% 2.3% 1.2% 2.3% 1991-2000 2.2% 0.6% 0.5% 1.1% 0.5% 1.O% 2000-2004 2.9% 0.6% 2.4% 0.0% 2.4% 0.0% Source: Authors' calculations from WDI data 5 inwhich TFP fell, although more recently, there is some indication that this is changing. While the human capital contribution is smaller, a characteristic of this methodology, the increases in human capital inIndonesia tend to be much larger. CumulativeTFP Once TFP is calculated for each year in the sample it is then possible to look at the cumulative evolution of total factor, or CTFP ,relative to some initial year, as in Solow (1957). Beginning with an initial value, such as 1, the TFP value for each year is added to the value of the Index from the previousyear as follows: CTFP, = 1, i f t = 0 CTFP, = CTFP,-, +TFP,, i f t > 0 Using this methodology, cumulative TFP is plotted in Figure 1 for Nigeria with initial capital- output ratios o f 1.5, and 2. For Figures 11-13, when comparing performance across countries, an initialcapital-output of 1.5 is usedby convention. From Figures 2 and 3 it is clear that Nigeria is not alone in terms o f lackluster total factor productivity. It has performed no better than Benin and Togo, although it has out-performed Ghana and especially Niger. On the other hand, as Figure 4 shows, there has been relatively little total factor productivity accumulation when compared against Indonesia. Still, Mexico, perhaps since the Latin American debt crisis of 1982, has converged back to a level o f TFP accumulated by the Nigerian economy. Venezuela has declined even more than Niger, although for probably very different reasons. 6 Figure1: CumulativeTFP for AlternativeInitial Capital-OutputRatios A 7 Figure3: TFP inNigeriaand Ghana and Togo - ................. Nigeria ----- Ghana Togo Source: Calculations ftom WDI data. Initial Capital-Output Ratio=l.5 Figure4: TFP amongsome OilProducingCountries lsvr"cs--/ / .................... .-/.< .......... ........... ...>--------. 9.. ........ ... / e A':. -/ .... ....... ................. ..... _.-..-.\, ... ...__.._ ...... -.-.-, - .................Nigeria ----- Indonesia Mexico Venezuela Source: Calculations from WDI data. Initial Capital-Output Ratio=l.5 8 TECHNICAL ANNEX 1-C:BENCHMARKINGNIGERIA TO OTHER OIL RICHCOUNTRIES I s Nigeria's response to oil price shocks much different from the way other Oil Rich Countries (ORCs) managed the windfall? Are there lessons to be learned from such a comparison?In this section we compare the size of the windfall, as well as the policy response to it, by Nigeria to a group of 22 oil-rich countries (ORCS).~Note that inthis section our measure of the oil windfall is imperfect as we approximate oil fiscal revenues with oil exports due to data availability; the problemwith this measure is that it represents gross ratherthan net oil revenues inthat profits of foreign oil companieshave not beentaken out ofthis measure. Despite the recent improvement in non-oil growth performance, Nigeria today is one of the countries most heavily dependent on oil amongst the 22 ORCs - it is actually third after Algeria and Angola? Creating broad-basedgrowth and diversifying the economy away from oil is still a key challenge facing Nigeria. Next, we compare the relative size of the oil windfall associated with the latest oil price hikes (2000-2005) for the 22 ORCs. The oil windfall is calculated as the cumulative annual increase in oil exports for the period 2000-2005 with respect to their level in 1999 (the last year of low oil prices). For the purpose of comparison, oil exportsare measured in constant 2000 $ billion. Nigeria's oil windfall during 2000-2005 amountsto $63 billion, more than 80 percent of its 2005 GDP, or $484 per capita, all measured inconstant2000 US dollars. Nigeria is fourth out ofthe 22 ORCs, by the size of the oil windfall, after Saudi Arabia, Russia, and Algeria. However, Algeria and Nigeria have a more extreme dependency on oil than any of the other ORCs, so managing this windfall is likelyto be most challengingfor these two countries. The oil-producingcountries inthe benchmarkingexerciseare Russia(RUS), Kazakhstan(KAZ), Nigeria (NGA), Indonesia(IDN), Azerbaijan(ME), Congo (COG), Algeria (DZA), Angola (AGO), Mexico (MEX), Equador (ECU), SaudiArabia (SAU), Iran(IRN), Oman(OMN), Bahrain(BHR), TirnidadandTobago (TTO), Cameroon(CMR), Venezuela(VEN), Kuwait (KWT), UnitedArab Emirates (UAE), Colombia(COL), Bolivia (BOL) andEgypt(EGY). Notethat to calculatethe relativeeconomic dependenceon oil we havechosen the share of oil exports intotal exports, as data for the share of oil fiscal revenue intotal fiscal revenue, as well as dataon the share of oil GDP intotal GDP are not readily available. 9 Figure 1: Oil export dependence and the size of oilwindfall during 2000-2005 Oil Worts % ofTotal worts2000-2005 amrage OilWindfall,2000-05 (constant2000 US$billion) I 250 ........................................................ ..................................... ................................................... ..................................... ............................... .......................... .............. Oil Wndfall, 2000-05 (% of2005 GOP) 011Windkll, 2000-05 (constant2000 US$,percapita) ................................................ 15000. 80 - .......................................... io000 - ....................................... 80. .............................. 40 - .................. 5000 - ................................................ 20 - ............. 0 - 0 - C K W ~ Z O U N Z O ~ ~ ~ W Z ~ U X > K Z - I - I ~ x m u m 0~ +c ~N ~ > u o ~~o z r ~ U ~ w w ~ Z~u~~ o - m o W o W s p O O 20 - .................................................. How have the ORCs used their recent oil windfall - how much was spent and how much was saved? And how much of the additional expenditure fell on investment rather than on consumption? Compare for the answer to those questions the cumulative increase in public expenditure (currentand capital) during2000-2005, againwith respect to their levels in 1999(the last year of low oil prices). The ORCs are then sorted by what amounts to a rough measure of their marginalpropensityto spendout of incrementaloilrevenues Colombia, Mexico, Egypt, Indonesia and Bolivia are excluded from Figure 1 because most of them are not heavily dependent on oil. None of them collectedan oil windfall of more than $18 billion, as can be seen in Figure 1. One reason for this is that these countries are not heavily dependent on oil - oil exports constitutes less then 25 percent of their exports and less than 5 percent of GDP (some of them, e.g. Egypt and Indonesia, have become net oil importers today). As a result, the enormous cumulativeincrease in public expenditure inrelationto the oil windfall in these countries may just mean that expenditure is being financed with predominantlyother sources of finance (e.g. non-oil revenues). Eliminatingthese five countries, Nigeria holds third place out of total 17 ORCs by the cumulative expenditure increases as a share of oil windfall. According to these calculations, so far Nigeria has spent about 70 percent of its oil windfall, which is morethan anybody else has done except Russiaand Camer~on.~ Note againthe caveat that at least a part of oil exports accrue to private sector (foreign or domestic). 10 Figure 2: Use of OilWindfall Cumulative Government Expenditure(% of Oil Wndfall) CumulativeSawngr,2000-05 (%of oilWndCII) 160 .................................................................... 250 ................................................................. ................................................................ ............................................................. .............................................................. ............................................................. ......................................................... .............................................................. ......................................... ..................... ...................................................... GobernmentEgenditure (% of Oil Rebenue) 160 ............................................................................ T 25 140 A 120 20 1w 15 80 ................ 60 10 - 40 5 20 0 0 2000 2001 2002 2003 2004 2005 ~ capitale g as % ofoil rebenua,w~th1999 as base ME4 currenteg as % ofoil rebenue,with 1999 as base +Oil Windfall,2000 US$billion. compared with 1999, nghtscala ButNigeriahas recently, in2004, implementedan oil price fiscal rule, intendedto de-linkpublic expenditure from volatile oil revenues. Therefore, we also check whether there is a change in spending pattern before and after the introductionof this rule. For this purpose, we present the share of annual increments in public expenditure (again with respect to their 1999 level) as a share of the incrementaloil windfall (with respect to the 1999 levelof oil revenues) of each year, so not cumulativethis time. The share of the oil windfall that has been spent has more than doubled in 2001 in relation to 2000 and was quite high in 2002 and 2003, but there seems to be a substantial drop in 2004 and 2005 - the years of the implementationof the oil price fiscal rule. Nevertheless, spending in percent of non-oilGDP increasedin 2005, but this increase was much smaller in relationto the oil windfall (as oil prices nearlydoubled in2005. Ofcoursewhether thisjust reflectsthe fact that thejump in oil prices took the authorities by surprise, liked in 1973/74, or whether a truly more conservative policy has set inremains for the future to tell; but at least there is a start of a more cautiouspolicy, which augurswell for a crash-less future. How has all this spending affectedthe economy?How severewas its impact on inflationandthe real exchange rate? To answer these questions we also plot average annual inflation and cumulativechange in the real effective exchange rate for the group of 17 ORCs (excludingthe countries with relatively low oil dependence). Nigeria was again among the countries that experiencedbothhighinflationand substantialreal exchangerate appreciation. 11 Iz5 11 Inflation( % 2000-05 average) $0 ..................................................................... ............................................................... ............................................................... ........................................................... 20 ..................................................... 15 .................................................... 10 ...................... 5 0 Note:The inflationrateinAngola was 135 percent on annualaverage during2000-05, which is not shown inthe figure Rankinghigh on the inflation and on the real appreciation scorecard is a strong indication that policy was inappropriate.It indicates that the nominalexchange rate was maintainedtoo rigidly, so that high inflation was the only way a real appreciation could come about. Other ORCs apparently have managed this better. Nigeria could either have taken pressure off the exchange rate by saving more, or should have allowed a nominal appreciation, so as to avoid inflation shootingup as a real exchangerateadjustment mechanism. Real EconomicGroWm (OA) Sectoral Composition +RealGDPp.c.GroWm(%,2000-05awrage) &riculture, mlue added (% ofGDP, 2000-04 a q ) Nigeria'sgrowthhas beenrobust duringthe past five years: Nigeriaoccupies 9* placewithin the sample of 22 ORCs. However, this ranking is somewhat distorted by steeply increased oil production in some countries (e.g. Azerbaijan and Kazakhstan). Finally, while Nigeria's growth has beenrobust, the share of manufacturingin GDP and exports is still the smallest in Nigeria, comparedwith other 22 ORCs. Summingup, Nigeriawas above average as to how much it spent and managedthe exchangerate consequences of the boom worse than average. Nigeria had one of the highest marginal propensitiesto spend, grew only at averages, and clearly followedan inappropriateexchangerate policyfor most ofthe time. 12 TECHNICAL ANNEX 1-D:CAUSES OF VOLATILITY INNIGERIA This annex details empirical analysis of the factors behind Nigeria's volatility. We have identified two different explanations for the high volatility, the voracity effect (VE) and debt overhang problems (DO). We saw that the "voracity effect ",an institutional inability to reconcile competing claims for oil money, results in a marginal propensity to spend out o f risingoil wealth larger than one, but a propensity to adjust inyears o f falling income that is smaller than one, and thus certainly smaller than the coefficient in "good" years. In effect, conflicts over resource use are resolved at the expense o f future generations. The DO hypothesis generates the exact opposite prediction, an adjustment coefficient in years of falling revenue that exceeds the coefficient in years of rising revenue. Thus what matters is NOT whether oil income is above or below trend, whatever the trend maybe, but whether oil income is rising or falling. This point is at the basis of the empirical analysis that follows. To test the VE hypothesis versus the DO hypothesis, we runa regression of government spending (also its components - current and capital spending) on oil-related fiscal revenues over the period 1970 - 2005. All data are presented in constant dollars. We first show the results for regressions with symmetric adjustment in up and down years (more accurately years with rising and years with falling revenues: (see Table 1below). Table 1: Government Spending and Oil Exports -Nigeria 1970-2005, levels in constant $ Total Government Current Capital Expenditure Expenditure Expenditure 1.45 0.87 0.52 Oil Revenue (9.87) (10.29) (6.36) 1.76 1.10 0.26 Constant (0.83) (0.90) (0.22) No. of Obs. 36 36 36 R-sq.overall 0.73 0.75 0.53 DW 1.13 1.46 0.73 Note: 1. Data source: IMFWorld Economic Outlook database.2. t-statistics inparentheses The regression shows a very strong relation between expenditure and revenues from oil, as is to be expected. The coefficient indicating the dependence o f aggregate expenditure on fiscal revenues from oil i s significantly higher than one. Moreover there seems to be a strong bias towards current expenditure, the relevant coefficient exceeds the coefficient in the investment only regression by more than two-thirds. But there is also a serial correlation pattern in the errors. The DW coefficient is too low in all three regressions, and Breusch-Godfrey Serial Correlation LM test rejects the Hohypothesis of no-serial correlation. Analysis o f the error terms furthermore suggests a before and after 1984 difference. The Chow Breakpoint test for 1984 strongly rejects the Hoo f no structural break in 1984 inall three equations. We therefore ran the regression including a dummy term for the period up to 1984. We also include the asymmetric adjustment coefficients: a dummy that equals one in years of rising revenue and 0 in years of falling revenue is entered multiplicatively with the oil revenue term. With this definition one should expect a positive coefficient under the Voracity hypothesis, and a negative coefficient under the D O hypothesis. Table 2 lists the results. 13 Table 2: Government Spending and Oil Exports -Nigeria 1970-2005, levelsin constant $ with OverhangVariable and 1984 dummy Total Government Expenditure Current Expenditure Capital Expenditure 2.01 1.19 0.79 OilRevenue (16.5) (14.87) (10.71) 2.52 1.20 1.01 Constant (1.86) (1.35) (1.23) Overhang -0.52 -0.3 1 -0.23 dummy (-4.66) (-4.24) (-3.47) Dum84*Oil -0.62 -0.30 -0.37 revenue (-5.78) (-4.22) (-5.72) No. of Obs. 35 35 35 R-sq.overall 0.92 0.90 0.84 DW 2.21 2.29 1.37 Note: 1. Data source: IMFWorldEconomic Outlook database. 2. t-statistics inparentheses The results again come out very strongly, but this time without indications of serial correlation (except for the investment equation). Consider first the regressiontaking total public expenditure as dependent variable. Before 1984 the spending coefficient in good years is still high at 1,49 (=2,01-0,52), and significantly higher than one. Thus at least before 1984, the fight for resources was clearly settled by shifting the burden out to the future, on the margin spending exceeded revenue. This changed in 1984, however. The 1984 interaction term suggests that the marginal spendingpropensity(msp)droppedby a substantialmargin, bringing the coefficient ingoodyears down below one, to 0,87. Thus to the extent there was a voracity effect before 1984, there is no sign of excess spending on the margin afterwards. Again there is a bias in favor of current over capitalexpenditure. Most interestingare maybethe coefficients on the overhangdummy, which shouldbe positive for the voracity effect to be dominant, and negative if the DO hypothesis is to be accepted as the dominant explanationof excess volatility. The overhang-dummyis negative and significant in all three regressions. So the combined coefficient during upturns (increases in oil revenues) is lower than the coefficient in downturns (decreases in oil revenues). Moreover, the value of the msp in bad years is clearly and significantly larger than one both for total spending and for government consumption, as the DO hypothesisprojects, while the msp in good years is below 1 in all cases, also in line with the DO hypothesis but at variance with the VE hypothesis. Thus the econometrics suggest a clear conclusion: there was some evidence of a voracity effect before 1984, particularly ingovernment consumption, but over the whole period and certainly after 1984 the Debt Overhanghypothesisdominatesstrongly. To see whether this is a Nigeria-specific problem or a more widely spread issue in all ORCs, we compare our results with the results from a panel regression for 22 oil-rich countries. However, since we generally do not have data for oil-related fiscal revenue, we have to use oil exports as a proxy for oil-related fiscal revenues. Once again, we regress public expenditure and their components on oil exports. 14 Sample: 1980-2005 RealOil exports = Oilexports incurrent US$dividedby U S GDP deflator Table 3: Government Spending and Oil Exports Oil-Rich Countries, levels in const $ - Fixed effects Fixed Effectswith OverborrowingDummy Total Total Government Current Capital Government Current Capital Expenditure Expenditure Expenditure Expenditure Expenditure Expenditure 0.52* 0.21 * 0.39* 0.73* 0.25* 1.09* Oil Exports (5.28) (3.95) (3.15) (5.24) (3.90) (6.63) 24.12* 12.18* 6.28* 23.26* 12.11* 3.19*** Constant (16.3 1) (13.43) (3.25) (15.22) (13.33) (1.65) Overborrowing -0.22** -0.06 -0.77* dummy (-2.11) (-1.07) (-6.17) No. of Obs. (No. 524 546 513 524 546 513 Of Groups) (21) (22) (22) (21) (22) (22) R-sq.overall 0.26 0.26 0.11 0.28 0.26 0.16 F Statistic F=27.89 F=l5.59 F= 9.90 F=16.27 F=8.37 F=24.36 (Prob > F) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) Note: 1.Data source: IMF World Economic Outlook database 2. t-statistics in parentheses; indicates significance at 1%, * ** at 5%, and*** at10% While we find that the coefficient on oil exports is positiveand significant in all three equations, it is much lowerthan one. When we addthe overborrowingvariable(see the last three columns), we find it negative and statistically significant in two of the three regressions (for total expenditure and for capital expenditure). The results therefore also suggest that overborrowing, rather than voracity, was the main issue inthe sample of 22 ORCs: there is clear evidence of a higher propensity to spend out of oil revenues in years of falling oil revenues than in years of rising oil revenues. However there is no ORC-wide indication of a spending coefficient larger than one in years of falling prices; thus the Debt-OverhangsituationNigeriafinds itself in seems more severe than obtains in other ORCs. The need for a policy response is correspondingly larger. Another interestingdifference is that in most ORCs there is a much tighter link betweencapital expenditure and oil revenues than in Nigeria. In Nigeria oil income goes mostly into consumption, in ORCs capital expenditure dominates. That is good in that assets are not just eaten up, there is an apparent attempt to use the asset draw down that oil extraction implies to build up other assets, in particular physical capital. The dark side of this medal is that the adjustment on baddays falls correspondinglyheavier on investment; the debt overhang dummy is negative and significantly so for capital expenditure and for total expenditure, but not for governmentconsumption. This is different inNigeriawhere the adjustment inbaddays hasto fall much more on consumption: most likely there simply is not enough public investment out of oil to implement the entire downward adjustment in bad days through cutting back capital expenditure. 15 TECHNICAL ANNEX 1-E: A FRAMEWORK FORASSESSING FISCAL SUSTAINABILITY Assessing fiscal sustainability for oil-rich countries requires distinguishing between the oil and non-oil fiscal positionn6Such a distinction is warranted because of the different nature of the oil- related fiscal revenue. First, oil is an exhaustible asset which means that fiscal revenues from oil extraction result from (natural) asset de-acummulation; this calls for treating oil revenue as a financing item, rather than current fiscal revenue. Second, faster oil reserves depletion today means that future generations would be worse off as oil will not last forever. To avoid leaving future generations worse-off, part of the oil revenue needs to be reinvested in other forms of assets/capital. Finally, oil revenue and implicitly the size of the oil wealth are volatile, mainly because o f oil price volatility; this complicates fiscal management and underscores the importance of accounting for oil price volatility when assessingfiscal sustainability. There are three important steps when implementing a sustainable fiscal strategy for oil-rich countries. Thefirst step i s to project the income stream of fiscal oil revenues, net of extraction costs, based on oil reserve estimates, oil price forecasts, and assumptions about oil exploration and extraction and oil sector taxation regimes. These projections should also account for the stochastic nature o f oil prices and for possible errors inassessingphysical oil reserves. The second step is to estimate the part of income flows that can be spent without depletingthe real stock of oil wealth: in an economists' jargon, the permanent income equivalent. Then having set the amount of resources available from oil revenues, a rule for the non-oil deficit path can be set up. Broadly speaking, this would depend on the rate of return to oil wealth. If the objective were to maintain the government's net asset position, then this income flow also defines the ceiling for the non-oil fiscal deficit. Ifthe non-oil deficit exceeds this limit, then the government would needto borrow, worsening its net asset position.' The third step requires modifyingthe government budget constraint and the resultingpublic debt dynamics equation to isolate the impact of oil on public finances. In particular, this requires modifications to reflect several major changes. First, it renders transparent the fact that a substantial share o f fiscal revenues is derived from oil; the primary fiscal deficit (non-interest spending minus revenues) is replaced with the non-oil primary deficit, isolating net oil revenues as a financing flow. Second, the change in net debt-to-GDP ratio now also accounts for fiscal savings out of oil, accumulated in a ring-fenced oil fund.8Third, giventhe importance of oil price volatility for assessing fiscal sustainability, it also deals explicitly with the impact of oil price uncertainty on net public debt dynamics. As a result, the net public sector debt dynamics is representedwith the following equation: d =b+b*-nfa*-oa* (1) = (f - Rref -a +( -g)6+( *-e- g)b*e-( *-e- ) r r r g)oa e - (Rweo - Rref) * + otherfactors 'Thisis section draws on PREMnote(2005) and ConceptNoteon How to ofFiscal Sustainability, mimeo 2005. This asimplified presentation. Borrowing to invest inhigh economic rateof returnprojectscouldstrengthenthe gublic finances over the long run. Inpractice, one has to contendwith wastefulspending andwhite elephantprojects. Ring-fencedoil funds can only be successful ifcomplementedwith arule that limits the non-oil deficit or public debt. Otherwise,thegovernment will accumulate assets in the oil fund while at the same time borrowing, so the net asset positionmay even deteriorate as the cost of borrowingis typically higherthanthe interestearnedon oil fund assets. 16 where Rweo refers to projected oil fiscal revenues (at projected WEODECPG oil prices); the oil price fiscal rule is captured by the projections for the non-oil primary deficit,J; which is financed by the oil revenue evaluated at the budget reference price o f oil, R r e j the change inthe net public debt ratio also accounts for the accumulationof assets ina ring-fenced oil fund, oa*-dot, which as equation 2 shows is equal to interest earning on the stock o f oil fund assets plus the oil revenue in excess to the revenue, evaluated at the budget reference price of oil; d, is the net public debt-to- GDP ratio, g is the real GDP growth rate, r is the real interest rate on domestic debt, r* is the real interest rate on external debt, rt i sdomestic inflation rate,rt* i s the foreign inflation rate, e-hat is the change in(bilateral, foreign currency per local currency unit) real exchangerateg. We then assume that the excess of actual oil revenue over the reference revenue evaluated at the budget reference price, and the interest earned on the stock of oil assets are saved ina ring-fenced fund: oa =(Y * *-e-/. g)oa e* +(Rweo- Rref) (2) Given that assumption, the gross public debt dynamics follow from equation (1). Applying the Framework: Is the Current OPFRFiscally Sustainable? Inthis section we use eq. (1) to assess the trajectory ofthe netpublic debtratio andthe robustness of oil price fiscal rule in the event o f a permanent oil price drop. As can be seen from equation (l),publicdebt-to-GDPratiocanchangeasaresultofchangesinnon-oilprimarydeficits,A the "automatic debt dynamics" (the difference between real interest and real growth rates), and changes incurrent oil revenues, Roil,related to changes in oil extraction profile, oil prices, and/or changes in fiscal terms for oil extraction sector. If a significant share of public debt is denominated in foreign currency, the public debt ratio can also change because of capital gains/losses due to real exchange rate appreciatioddepreciation. Finally, other factors, such as contingenuhidden fiscal obligations, can lead to public debt accumulation. Not surprisingly, public debt dynamics are particularly sensitive to assumptions about projected and budget reference oil prices, which determine the net oil revenues and acceptable level of non- oil deficits. As we have seen, the cost of oil price forecasting errors are asymmetric, with the cost of over predicting being far greater than the cost of under predicting. To account for this feature, the annual budgets for 2004 and 2005 were built on rather conservative budget reference oil prices. Looking forward, budget reference price is assumed roughly constant at slightly above $30 per bbl, despite expectations that oil prices will stay at their current high levels for the projection period. Given the current WE0 projections of a fairly high oil prices, hovering above $60 per bbl, this will result ina substantial savings out o f oil revenues. e-haP0 denotes a real exchange rate appreciation. 17 Figure la: Oil Price and Figure lb: Net Oil Revenue Oil PnceScenarios, UShbl 50-................................................................ NetOilRevenue,USGhillion 25 I S , I I , , , , , , , , , , , 1999 2000 2001 2002 2001 2004 2005 2006 2007 2008 2009 2010 2011 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 --cIMFWE0pnce --tBudgetPrice --cat IMFWE0 price t a t Bud@ Price All the assumptions underpinningthe fiscal sustainability calculations inthe base case are derived from the IMF projection framework. Fiscal projections assume that non-oil tax collection will improve and the non-oil taxes will increase by about 3 percentage points of GDP. Public expenditure is assumed to increase but by less than the increase in the non-oil tax revenues, so that the entire non-oil deficit i s being financed by the oil revenues calculated at the budget reference price. The non-oil deficit to non-oil GDP ratio is on a declining path, reaching 35 percent in the medium term, which according to the IMF i s consistent with maintaining real wealth over the long run. In addition, the base case scenario assumes that under good policies and reforms, the non-oil GDP grows at about 5 percent, underpinned by a strong growth in the non-oil sector (on average constant - at around 4 percent as a ratio to GDP, with very low real interest rate on external debt, 6 percent over the projection period. External debt to GDP ratio is assumed to remain roughly reflecting Nigeria's low-income status. The annual average real interest rate on domestic debt during the projection period is assumed to be somewhat lower, 6 percent, accounting for the ample liquidity and increased savings from oil. The base case also assumes no significant appreciation or depreciation inreal exchange rate, which once again reflects the relatively prudent fiscal stance under the current assumptions that highoil prices will prevail. Table 1: BaselineMacroeconomicAssumptions and Debt Sustainability Baseline Scenario 2005 2006 2007 2008 2009 2010 2011 Oil Price US$/bbl IMFWE0 average price 54.2 60.5 58.5 57.5 56.9 56.4 56.1 Budget referenceprice 30.0 33.0 30.0 30.0 30.6 31.2 31.8 Real GDP Growth PA) Oil GDP Growth 5.5 5.1 5.1 5.4 4.0 4.0 4.0 Non-OilGDP Growth 8.2 7.0 5.3 5.4 5.5 5.5 5.5 OverallGDP Growth 6.9 6.1 5.2 5.4 4.8 4.8 4.8 RealExchangeRate(%change, + depr.) -15.9 -4.9 -0.2 0.7 2.0 3.2 3.2 Inflation Rate (%) 19.2 11.5 6.5 5.5 4.2 3.0 3.0 Real interestrate (%) 6.3 2.6 5.8 6.4 7.2 7.8 7.1 Non-oil primary deficit, YOof GDP 19.1 17.9 18.4 17.1 17.5 17.7 18.5 Non-Oil PrimaryDeficit, YOofNon-Oil GDP 41.4 38.1 38.5 35.0 35.0 35.5 36.5 Oil revenueat reference price, % of GDP 18.4 19.9 19.3 18.8 19.2 19.3 18.8 Non-Oil Revenue, %of GDP 6.6 7.9 8.4 9.0 9.6 9.8 10.0 Oil stabilizationfund net assets, %of GDP 28.5 46.9 64.7 81.5 96.1 110.2 123.7 Gross PublicDebt, %of GDP 31.0 16.8 15.4 14.3 13.7 13.1 13.0 Net PublicDebt, % of GDP -2.2 -23.7 -41.8 -59.7 -76.3 -93.2 -108.3 As seen from 2, if the oil prices remain at their present very high levels of around $60+ per bbl, fiscal sustainability i s not going to be an issue in Nigeria for many years to come if the current OPFR is adhered to. Following the oil price fiscal rule would lead to a very fast accumulation of foreign assets: at the end of the projection period the net asset position will reach 100 percentof GDP! Sizeable oil assets accumulation is the chief reason for this favorable debt dynamics. Of course, the strong initial position (as a result of the combination of the Paris Club debt relief and the highoilprices) also helps. As demonstratedby Nigeria's past, the role of fiscal policy inminimizing the impact of oil wealth volatility on the economy by de-linking public expenditurefrom volatile oil revenues is crucial if debt overhang and associated low growth are to be avoided. To check robustnessof the OPFR, we also project the net public debt dynamics on the assumption of a negative permanentoil price shock during the projection period. Inparticular, we assume that the oil price would permanently drop to the average oil price during 1986- 1999, or to a constant $25 per bbl inreal terms. At the same time, we also assume that fiscal policy is still set according to the fiscal rule, which stabilizesthe non-oil deficit at about 35 percentof non-oil GDP inthe medium term. Under this scenario, a return of the oil price to historical averages, the budget reference price is slightly higher than the assumed oil price under this scenario ($25 per bbl in constant prices). This would lead to net outflows from the oil stabilization fund, to compensate for the difference betweenthe budgetreference price andthe actual oil price, but the fund would not be exhausted. Thus if the recently adopted fiscal strategy is continued, and given the sound initial net asset position, it is likely that Nigeria would be able to survive a sizeable negative oil shock of at least another 5 years duration, without requiring dramatic fiscal adjustment and also with only modest debt accumulation. This is good news for the fiscal rule and indicates that it is robust enough to withstand a sizeable negative oil shock. Of course, if the duration of the shock is much longer (beyond the 5 years projection period which we assumed for our purpose), this policy rule may needto be reconsideredto avoid overborrowing. Table 2: OilPrice and Debt Sustainability - Lowest Oil Price Scenario 2005 2006 2007 2008 2009 2010 2011 Oil Price, US$/bbl 54.2 25.0 25.5 26.0 26.5 27.1 27.6 Oil stabilization fund net assets, % of GDP 28.5 24.4 21.3 18.6 15.2 11.9 9.0 Gross Public Debt, % of GDP 31.0 16.8 15.4 14.3 13.7 13.1 13.0 Net Public Debt, % of GDP -2.2 -1.2 1.0 1.9 3.4 4.8 7.2 Finally, we have constructed an alternative scenario, where we explore what would be the consequences of abandoning the fiscal policy rule and spending according to current high oil price projections, while in reality oil prices would drop permanently to constant $25 per bbl. In other words, we perform a robustness check to see what would be the implications if history repeats itself and Nigeria would experience a similar shock as the one of the mid-1980s. Specifically, this scenario assumes that the government will passively spend its oil revenues, which are estimated assuming that current high oil prices ($60 per bbl) will continue to prevail during the projection period , while oil prices in fact returnto their historical "base level" ($25 per bbl inrealterms). This would lead to rapid increases in the non-oil deficit, while at the same time oil prices may actually return to the historical lows and the oil revenues available to finance those deficits turn out to be much lower than expected. Under such a policy, Nigeria will exhaust its oil fund assets very rapidly in 2006, after which sizable financing gaps emerge, as shown as the difference 19 between projected non-oil primary deficits and oil revenues. The options available to the government are then to borrow or, more likely when external funds dry up, to print money and see inflationre-emerge. Table 3: Oil Price and Debt Sustainability Country Specific Scenario- 2005 2006 2007 2008 2009 2010 2011 Oil Price, US$/bbl 54.2 25.0 25.5 26.0 26.5 27.1 27.6 Budget reference price, US$/bbl 60.0 60.0 60.0 60.0 60.0 60.0 Real GDP growth (%) 6.9 6.1 5.2 -6.0 -5.0 -3.O 0.0 Real interest rate (YO) 2.6 5.8 20.0 15.0 10.0 5.0 Real Exchange Rate ("A change, + depr.) -4.9 -0.2 25.3 2.5 2.5 2.5 Inflation Rate ("A) 11.5 6.5 80.0 40.0 20.0 20.0 Non-oil primary deficit, % o f GDP 19.1 34.0 32.4 45.3 48.1 52.1 52.1 Non-Oil Primary Deficit, % o f Non-Oil GDP 41.4 74.7 71.4 152.6 196.1 228.3 252.4 Oil revenue at reference price, % o f GDP 18.4 34.0 32.4 45.3 48.1 52.1 52.1 Oil stabilization fundnet assets, % o f GDP 28.5 0.0 0.0 0.0 0.0 0.0 0.0 Gross Public Debt, % o f GDP 31.0 17.5 34.6 64.8 99.9 137.7 169.0 Net Public Debt. % o f GDP -2.2 17.5 34.6 64.8 99.9 137.7 169.0 Given the debt overhang problems for oil-rich countries after price downturns, this would most likely leadto confidence crises, creditrationingby external capitalmarkets, sizeabledepreciation, rapid increases in real interest rates, hyperinflation, and deep real economic recession, the end resultofwhichwouldbe againunsustainablelevelsofpublic debt (mainly domestic) only 5 years after the PC debt agreement. This simulationthus stressesthe importance of the fiscal policy rule andthe conservativereference pricechosento underpinit:bothare necessaryto impact sufficient robustness to Nigeria's macroeconomic policies: guessing wrong can be very expensive if the mistake is assuming high prices and adjusting expenditure accordingly,while in fact low prices emerge. Figure 2: Public Financing Gap in the Country Specific Scenario 60 -.............................................................................................................. 50 .................................................................. ............................. - 40 ............................. ................................................................ 30 - ................................................................................................................ ------- 20 ...........- .. .L. .u. ...................................................................... -4.. .e.. 110 +......................................................................................................... 2006 -Non-oil 2007 2008 2009 2010 2011 --Oil primary deficit, % of GDP reenue awilable for spending, YOof GDP 20 TECHNICAL ANNEX 1-F:PRODUCTIVITY AND COMPETITIVENESS IN AGRICULTURE Given the importance of agriculture in the non-oil economy, current low and declining level of productivity in agriculture is a significant threat to sustaining Nigeria's recent non-oil growth. While impressive, growth in agriculturein the last few years has not been driven by productivity improvements and is therefore unlikely to be sustained over the medium term. Improved agriculturalproductivity is essential for transformingthe sector from subsistence to commercial agricultureable to support a modern agri-business industryand growing non-oilexports. This is essential for successfully diversifying the economy away from oil. Moreover with low agricultural productivity,job creation in the rural economy and in related industry will remain limited both in numbers and in levels of remuneration. Nigeria's poverty reduction objectives will be harder to achieve. Raisingagriculturalproductivityand competitiveness therefore needs to be a central elementofNigeria's effortsto achieve sustainedgrowth. Several sectors specific issues that combine with distortions in the business environment and weak access to infrastructure (discussed in the preceding chapters) to keep agricultural productivitylow needto be addressed. Intacklingthese issues the government will also needto reviewthe broader issue ofpublic sector supportto agricultureand seekto becomemore strategic in approach. The next section analyses the main sector specific constraints to agricultural productivity; section two comments on key aspects of the government's approachto support for agriculture;sectionthree providessuggestionsfor addressingthe key sector specific constraints to productivityandkey recommendations are summarizedinthe last section. I.FactorsinLowproductivityandcompetitiveness Productivity has declined for both commercial and food crops in Nigeria over the last twenty years. For commercial crops production levels have fallen as well (provide evidence). In contrast, productionlevels of food crops have increased substantiallyand Nigeriahas overcome the extreme import dependencefor basic foodstuffswhich occurred inthe 1980s. However, this growth has been driven by steady and substantial increase in the area cultivated and harvested" while yield levelshave dropped. Inthe case of roots and tubers for example, a fourfold increase in area planted since the mid 1980s has beenaccompanied by a decline in yields in excess of 40 per cent. Similar, but less dramatic outcomes are observed for cereals, beans and groundnuts. (see Table 1)" This provides strong evidence that a policy of targeting output based on area expansion is inappropriate.Diminishingreturns to land appear to have clearly set in a longtime ago. Table 1: Nigeria: Main Crops, Area Harvested, Production and Yield, 1999,2003 and 2004 Area Harvested Production Yield 1999 2003 2004 I 1999 2003 2004 I 1999 2003 2004 MainCereals 18437 24133 25515 22478 29186 31477 1.22 1.21 1.23 Roots & Tubers 6849 9300 10344 66076 78830 84465 9.65 8.48 8.17 Pulses 10340 13961 14107 4603 5952 6140 0.45 0.43 0.44 Total 40920 54574 56753 lothe statistics relateto area harvested and are likelyto understatearea cultivated since areas which yield no harvest are not included Itshouldbenotedthat thetable excludes the period 1972to 1985for which comparable informationis not available, notably for areaharvested and yields. 21 The fundamental cause of low agricultural productivity in Nigeria is the very low use of modern technology evidenced inweak research and extension, limiteduse o f improved seed varieties and breeds and of irrigation. Inaddition, weak human resource and skills base are also contributory factors. Research. Nigeria's national research system has enjoyed only limited success ingenerating new technologies that have been taken up by farmers. This i s due to (i)poor funding of public research organizations; (ii)weak coordination within the Nigerian agricultural research community, resulting in unnecessary duplication of effort; and (iii) tendency for research to be supply-driven, with little accountabilityto farmers. Public institutes responsible for conducting agricultural research inNigeria have historically been underfunded. Since 1992, responsibility for overseeing agricultural research has been vested in the Agricultural Sciences Department (ASD) within FMARD. Today, ASD coordinates the activities of the 15 NARIs located throughout the country, but it has neither the financial resources nor the implementation capacity to carry out its mandate. Within the NARIs, budgets have remained flat even as staffing has increased, forcing severe cutbacks in operating budgets. Lack of systematic collaboration between research institutions in the sector has created sub- optimal allocation o f resources characterized by duplication of effort in some areas and underinvestment in others. Farmers have had limited influence over the orientation of research, leadingto the development of technologies that do not address farmers' problems. Extension. Extension services in Nigeria are delivered mainly through public agencies. A handful of private agribusiness firms, mainly input dealers, provide extension advice as a service to their clients, but the coverage is limited to a few crops. A third source o f extension services i s NGOs, community-based organizations, and producer organizations. Producer organizations are growing in number and playing an increasingly important role in fostering farmer-to-farmer technology transfer. Public agricultural extension programs in Nigeria are vested in all three levels of government. The weakness of the extension system is primarily due to chronic underinvestmentwithin each level as well as poor coordination between levels. Similar to the research system, the extension system lacks accountability to farmers. Starting in the 1 9 8 0 the ~ ~ federal government established a unifiedagricultural extension system. Although coordinated at the federal level, this system is implemented through the ADPs, which are run by the State Ministries of Agriculture and Natural Resources. At the local level, many LGAs maintain agricultural units that offer extension services. Recently, all states agreed to standardize extension service delivery through the LGAs. The common approach has met with mixed success, with considerable portions of the system largely dysfunctional, the ADPs have de facto remained the main public provider of extension services and the quantity and quality of extension services providedto farmers continues to be inadequate. Low use of improved plant varieties and animal breeds. The adoption of modern varieties ( M V s ) and improved breeds in Nigeria i s extremely low. M V s of most leading food crops are available through many ADPs, but the area planted to M V s is still modest (see Table 2 below). Most farmers still plant traditional varieties using seed and planting materials saved from their own harvest or obtained from local sources, including relatives, neighbors, or local traders. 22 Table 2: Area Planted to Modern Varieties of Major Food Crops, Nigeria, 2000 Crop Area cultivated Estimated area planted to MVs (million ha) (YOof area planted) Millet and Sorghum 3.71 30 Maize 2.93 20 Cassava 2.09 50 Yam 1.94 10 Rice 1.59 30 Source: FGN2003 The low use o f M V s is explained by both supply and demand-side factors. The single most importantsupply-side constraint has beenunderinvestmentin crop improvement research. Public plant breeding programs inNARISare understaffed and underfunded. Privatecrop improvement research is almost non-existentin Nigeria, because commercial seed marketingopportunitiesare few and far between. Opportunitiesto import M V s developed outside the country are limited, either becausemany of the crops grown in Nigeria are "orphan crops" that attract little attention from breeding programs in other countries (e.g., yams, cassava, millet), or because production conditionsinNigeriaare sufficientlydistinctivethat extensive localadaptationbreedingwould be needed(e.g., maize, sorghum, rice, wheat). Low effectivedemandis also a factor. ManyNigerianfarmers have never beenexposedto M V s and are unaware of their potentialbenefits. They cannot affordthe MV seed or the fertilizer and other inputs needed for M V s to performto their full potential. Finally, the additionalbenefits may not justify the additional costs of M V s especially, in marginal production environments where weather-inducedcrop lossesare common. Low use of purchased inputs. In the crops sub-sector, the use of fertilizer and pesticides is very low. In 2000 the amount of fertilizer appliedto field crops represented only about 3 percent of the totalagronomic requirements (FMARD2004). A major factor is the highcost, which, inturn, results from low levelsof importsand limiteddomestic production. Highcosts also reflect poorly developed markets, low quantities transacted, high transportationcosts, and transactions costs associated with regulatory compliance and sometimes corruption. For many imported inputs, internaltransport and handlingcosts are greater than internationaltransport and handlingcosts. For example, moving fertilizer from Lagos to Abuja adds $50 per ton to the cost, and moving it from Abuja to the north of the country adds a similar amount12 These large internaltransport costs contributeto some ofthe highestretail pricesfor fertilizer inthe world (see Table 3). Limited Area under Irrigation. Agricultural productivity is low in Nigeria in part because agriculture is risky. Investorsare reluctantto commit resources to activities whose returns are dependent on highly variable rainfall patterns. Irrigationcan increase and stabilize yields and production, and contribute to productivity growth by improving the incentives to invest in improvedtechnology and inputs. The area under irrigationwill needto expand at unprecedented rates to allow irrigation make a significant contribution to agricultural growth. Estimates of Nigeria's irrigation potential range from 1.6 million ha (FA0 1991) to 2.5 million ha (FA0 2000b). Currently only about 220,000 ha (0.7 percent) of the nation's cultivated land is under irrigation. Low profitability of irrigated crop production, poor public management of water resources, andweak privatemanagementcapacity are key factors. ''IFDC 2005 23 Expansion in the area under irrigation will occur only if irrigated crop production is demonstratively profitable for farmers. Currently, participation in existing large- and medium- scale public schemes i s financially attractive to farmers only becausethe capital investment costs of irrigation infrastructure are paid by government. If the capital investment costs of these schemeswere charged to farmers, irrigated crop productionwould no longer be pr~fitable'~ . Table 3: Urea cost in selecteddevelopingand transitional countries, 2001 Bangladesh 221 Brazil 304 Thailand 327 Poland 370 Nigeria 566 Source: FAO, IFC Most public irrigation schemes inNigeria are poorly managedI4. A recent review of 62 schemes found well over 50 percent in complete disrepair, while most o f the rest were operating at low levels of capacity. Poor public funding and inadequate cost recovery mechanisms are important constraints. The situation i s aggravated by institutional deficiencies. The technical and management capacity of most River Basin Development Authorities (RDBAs) is weak. Inmost instances, water user associations ( W A S )either do not exist or are ineffective. There i s no clear interface and only weak coordination between the Federal Ministry of Water Resources (FMWR) and FMARD. The factors described above, have combined to exacerbate poor soil quality of oil. Ingeneral, while Nigeria has a relatively highproportion of total area which is cultivable, the quality of the soil is typically not good. The rapid increase in cropped area, which has in many areas been achieved by reducing the period of fallow and reduced opportunity for fertilization through animal grazing inthe off-season has reduced the ability of the soil to recover. Weak Human Capital Base. Productivity growth in Nigerian agriculture will not be achieved without improvements inthe human capital base for agriculture. The human capital base in rural areas is constrained by social and demographic factors such as: (i)low farmer education; (ii)an aging farming population; (iii)the marginalization of women in agriculture; and (iv) health- related productivity losses attributable to malaria and HIV/AIDS. Educational access and quality in Nigeria is low and rural schools are under-fundedrelative to urban schools and attendance and completion rates for rural schools lag behind those for urban schools. Low profitability and heavy physical labor demands have combined to make agriculture unattractive as a livelihood source. The resulting exodus o f younger people to urban areas has createda pronounced aging o f the farming population (what is the average age of the Nigerian farmer?). Recent surveys confirm the presence of a growing pool of surplus (unemployed) labor in the cities, at the same time as tightening season labor markets in rural areas. The well- established patterns o f seasonal migration seeking agricultural work mean that there is often a conflict between migration and availability to help with peak season requirements at home. All of these factors tend to raise the cost of labor in rural areas especially inthe north of the country where many of the food staples are dominant. Women play a pervasive and central role in Nigerian agriculture, yet they face a number of barriers that reduce their contribution to agricultural growth. They have greater difficulty accessing land, labor, and capital. With l3 PSE2002 l4(FA02004) 24 extension services biased toward serving men, women also typically face restricted access to technical information. Illness is another major constraint to agricultural productivity growth. Malaria has been identifiedas the most prevalent disease inNigeria, followed closely by waterborne diseased5, In a study of severely affected countries, including Nigeria, malaria was found to reduce economic growth by 1.3 percent annually16. HIV/AIDS has a lower incidence rate in Nigeria than either malaria or waterborne diseases (approximately 5.1 percent o f the population was afflicted with HIV/AIDS in2005), but the economic costs are higher over the longer term. Malaria, waterborne diseases, and HIV/AIDSexact a heavy toll and diminishthe productivity of the agricultural labor force. These and other diseases constrain productivity growth in agriculture, both directly (by reducing the quantity and quality of available labor) as well as indirectly (by diverting financial resources for treatment). Low productivity from the factors discussed above, combine with other macro policy including trade distortions" to make Nigerian agriculture highly costly and uncompetitive. 111.GovernmentApproachto AgriculturalSupport Agricultural policies and programs in Nigeria have evolved through a series of distinct phases'*. The colonial period and the first post-independence decade was characterized by an export- oriented, laissez-faire approach. The 1970s up to the mid 1980s was marked by massive government interventions in the agricultural sector. Between 1985 and 1999 implementation of structural adjustment program led to government withdrawal from direct participation in production activities. Since 1999, the emphasis has been on the creation of the enabling environment neededto encourage greater private investment inthe sector. The evolution of government approach to agriculture over the last three decades is broadly inthe right direction-allowing private sector to lead. However a common feature running across all four phases has been a heavy reliance on special initiatives, programs, and projects to provide immediate solutions to pressing crises. Against a background o f short political cycles, agricultural policies have tended to change frequently with changes in political leadership. Furthermore, many special initiatives and programs, although well-intentioned, have created more distortions, have been financially unsustainable, contributing further to the discontinuity in policies and programs. The fragmented approach to policy-making and specific interventions has prevented emergence of a coherent, integrated strategy for effective government support to Nigerian agriculture. Table 4 below provides a summary o f key incentives to promote investment and exports inagriculture. Policies have also often been poorly coordinated across tiers of government. Federal, state, and local government agencies often share responsibility for similar functions, but in many cases fail to coordinate effectively. The lack of effective policy coordination within the agricultural sector has often been confounded by poor coordination between agriculture and other sectors. For example, even though FMARD and the State Departments o f Agriculture have consistently emphasizedthe importance o f improving market access, the Federal Ministryo f Transport has not always made needed investments in rural roads, limiting market access. Similarly, agricultural and rural development policies have consistently emphasizedthe importance of promoting value- World Bank2001 l6 Sachs 2001 Trade Policy Issues are discussed in chapter 6 World Bank 2001, Okuneye2005 25 adding activities as the basis for growth. However, the strong urban bias in national power development strategies-- the availability o f electricity in rural areas is less than half than that of urban areaslg--has constrained the development of post-harvest processing inmany rural areas. Table 4: Summary o f Government Incentives for Accelerating Investment and Export Performance in Agriculture S/N Incentives Objectivekomments and status of Implement action 1 Refinancing and Rediscounting To provide liquidityto banks in support of their finance business . - _ _ Faculty (GF) and Foreign Input directed on export promotion and development. RRF took off in Facility (FIF). 1987and FIF in 1989 2 Currency Retention Allow exporter to hold export Scheme proceeds in foreign currencyin their bank. Took off in 1986. 3 Tax Relief on Interestsearned by To encourage banks to finance exports by reducing their tax Bankson export credit burden. Becameeffective in September 1986 4 Export Credit Guaranteeand Assists banks to bear the risk in export business, thereby InsuranceScheme facilitating financing and export volumes 5 Duty Drawback Scheme To reimburse custom duty paid by exporters on imported input used for Banks and CBN export production. This has Been widely utilized by exporters due to The cumbersome procedural requirements involved, although the fund was in 1988 increased to N5Omillion. 6 Export Expansion Grant To encouragecompaniesto engage in export businessrather than domestic business especially exporters who can export N50,000 worth of semi manufacturedor manufactured products. 7 Export Price adjustment scheme This is a form of export subsidy designed to compensate exporters of products whose foreign prices become relatively unattractive to exporters due to factors beyond the exporters control. 9 Export Development fund To assist exporters in partly paying the costs of participation in trade fairs, foreign market research, etc. This is an old scheme. - 12 Accelerated depreciationto capital To extend supplementary incentive industrial organizations for allowance export oftheir products-starting in 1986 13 Manufacturing-in-Bond Scheme To assist potential exporters of manufactured products to import duty- free, raw materials for production of exportable products 14 Presidential Initiatives A wide rangeof sub-sectorbased initiatives since 1999 to achieve acceleratedinvestment and growth through a combination of assuredprotection combined with other support instruments, basedon active PPPs. Sectors with ._ initiatives include cassava, fish, rice and hrniture. Source:Adapted from: Ikpi(2002) The fertilizer support program for example has been distortionary, inconsistent and poorly coordinated. As part of early industrialization efforts, the Federal Government invested heavily C. Willoughby 2004 26 in fertilizer plants (National Fertilizer Company of Nigeria in Port Harcourt and Federal Superphosphate ManufacturingCompany in Kaduna). These were eventuallyclosed down since they were not profitable. At the same time a fertilizer subsidy (25% of import parity price) was introduced through the Fertilizer Market Stabilization Program (FMSP) to improve the affordabilityof importedfertilizer. Most of the fertilizer distributedthroughthe FMSP ended up in the hands o f large-scale commercialfarmers with political connections and did not reach the small-scale farmers for whom it was intended. In addition, the subsidy had the opposite effect from the intended effect of creatingincentives for the emergence of a private fertilizer industry. Since it undermined incentives for private suppliers to invest in fertilizer procurement and distribution. Because ofthe highfiscal cost, the subsidyprogramwas eventuallyscaledback, but privatefertilizer suppliers have not beenable to step in and fillthe vacuum left by the withdrawal ofthe public sector. Two more recent specific support programs are the PresidentialInitiativesin Agriculture and the Financingfor AgricultureProgram(box 1below). Severalimportantlessonscan be gleanedfrom pastexperienceandusedto reformthe current approachto support inagriculturein order to ensurethat productivityand competitivenessis enhanced. The most importantare discussedbelow: First, stand alone specific interventionsandprograms cannot substitute for broaderpolicy reforms to underpinproductivityimprovements. Ina distortedgeneral environment, it is unlikely that these specific interventionscould have the maximum impact and the costs of achieving any impact are likely to be unacceptable. There is a danger that concentration o f policy attention on specific initiatives could divert attention from addressing the more fundamental policy and institutional constraints in the business environment facingagriculture inNigeria. For example, progress inaddressingthe seeds issue inthe rice sector under the PresidentialInitiativesshouldnot detract from the wider objective of re-activating the institutional structure for seed development, breeding, multiplication and certificationfunction for all crops. Inaddition, in a distortedgeneral environment, it is unlikely that these specific interventions could have the maximum impact andthe costs of achievingany impact are likelyto be undulyhigh. Interventionsare likely to be more effective when they are targeted broadly rather than narrowly. There are two reasons: it is usuallyquite difficult especiallywhen the choice i s beingmade by government to identify the correct target be it a subsector or produc?'. Secondly, and perhaps more importantly, what matters more for productivity and competitiveness of agriculture is the creation of value along the supply chain of a particular "secto?' " Value, is realized by the buyers along each node of the chain. Targeting one component of the chain, e.g. producers of rice, as opposed to tackling constraints confrontingthe value chain as a whole, includingthe suppliers of key inputs and services to producers; firms that processrice; wholesalers and retailers; and of course consumers, will likely have a limitedor disappointingoutcome. The current ban on rice importscanbeusedto illustratethis point(Box 1below) ' OEspeciallyinsituations where there is incompleteinformationandwhere circumstancesmay change very quickly "wheresectorisdefinedasthegroupofunitsproducingcompetingproductsandservices 27 Box 1: Rice Import Bans Banning imports has raised prices to consumers while on its own the ban is unlikely to stimulate significant productivity and investment in the "rice industry". A broader view of the rice value chain can help identify a range of measuresthat together could prove highly effective. Some are longer term, e.g. a reliable electricitysupply for processors,or the development of appropriate"packages", including improved seedvarieties, methods of cultivation, and fertiliser andpesticidetreatments for different size producersinvarious parts of the country. Some are more immediate: removal of unnecessarybarriers to the inter-state movement of goods, strategic public orjoint publidprivateinvestmentsin storage and distribution; and expedited delivery through the ports of imported equipment for processing and packaging. Directedor subsidisedlendingmay not be required,but rather innovationsinfinancing, e.g. equipment leasingand supplier credits. These can be identifiedthrough exchanges, possibly leadingto partnerships, among governments, businesses and financial institutions. Within this broad array of measures, a banonrice imports may be found to be ineffectual or even counter-productive inachieving the desiredoutcomes. Incontrast, a well defmed"smart intervention" wouldyield tangible benefitsfor many Nigerians, and gain their support for other less direct, more protractedbut very important policy measures. It would also be importantthat interventions not further accentuate the duality in the sector and further accentuate the disadvantagethe very large number of informalsector processorsof primary products who constitutethe larger part of the agribusinesssector as a whole face. Ways must be found to encourage smaller informal operators to increase their integration into the formal agricultural processingandmarketingchains. 0 A third lesson is that support programs need to be designedto emphasize accountability and results. This is important to ensure that preferential treatment through these programs is made conditional on achieving clear results including output, exports were relevant etc. They need to be designed within frameworks to ensure that results are delivered and that these are carefully monitored and evaluatedto contributeto the overall growth objective. 0 It is important to carry out cost benefit analysis of these different programs to establish that they represent value for money inthe use of public resources and to provide a basis for selection betweendifferent options that may be open to Governmentto select from. 0 Finally attention is neededto ensure that these special initiatives/programsdo not leadto a re-drawing of the lines defining the rolesof the public and private sectors in agriculture. Inparticular inthe areaoffinance for the sector, theroleofthe public sectorshouldbeto provide the environmentwithin which there is an adequate incentive for the private sector to invest. This framework should also include a clear plan for phasing out such special programs or for their eventualtake over by private sector where relevantand feasible. IV. AddressingLowProductivityand CompetitivenessinNigerianAgriculture In addition to addressing the broader business environment and infrastructure constraints discussed inthe last two chapters, government also needs to address the sector specific factors in weak agricultural productivity and competitivenessdiscussed above. Reforming research While a considerable new technology can be imported from outside the country and adapted to local circumstances, the distinctive features of many of Nigeria's production systems mean that some technology will have to be generated domestically. Research must be reformedto achieve 28 this. In particular, NARIs must be strengthened and made financially viable. Funding for research must be increased, and the focus shiftedto becomemore demanddriven. Improving the Institutional Sustainability of NARIS. Reformsto the nationalagricultural research system must begin at the federal level. The issues of institutional restructuring, ensuring sustainable financing, improving management, and ensuring the accountabilityof researchers to clients and funding agencies need to be addressed. All research conducted by public agencies (includinguniversities) should take place inthe context of a nationalagriculturalresearch policy, designed to reflect the priority needs of farmers and agribusiness enterprises. Nigeria should implement some pilot projectsto reform research institutes in terms of governance, legal status, and management. Policiesto encourage increasedprivate sector research should be put in place, including policies to ensure an appropriate regulatory environment (e.g., intellectual property rights and biosafety). Regulations that unnecessarily restrictthe activities of private firms (e.g., seed laws) shouldbe revised. Box 6 describesBrazil'sexperienceinthis regard. Improvingfunding for Agricultural research. Agricultural research in Nigeria, as in many other developing countries, must be supported by public investment. The Nigerian government's fundingfor agriculturalresearch, currently about 0.02 percent of GDP, has in recent years fallen well below the average even for Africa (0.85 percent of GDP). Given the unreliability of government support, mechanisms used to fund agricultural research in Nigeria should be diversified, and the source of funds expandedto includepublic, private, and beneficiarysources. The following funding options merit consideration: (a) Contractual funding could be used for priority research; (b) Competitivefunds could be used to finance some high-priority national research activities, especially upstream research, and research within innovative partnerships among NARIs, universities, and the private sector. Use of competitive funds is feasible in Nigeria. One source of funding for agricultural research could be NEPAD; (c) Userfunds could be piloted for adaptive research. Some of these funds should be channeled through producer organizations for priorities defined by the organizations themselves. To create a demand-led system, farmers must be put in the driver's seat, so producer and community organizations must be empowered to participate effectively. Traditional approaches to organizing and forming cooperativesshouldbe revisedto meetthese newdevelopmentchallenges. A revitalizedand strengthenedAgricultural ResearchCouncil of Nigeria (ARCN) should be the major body for setting research policy, coordinatingresearch, and funding research throughout the country. Its membership should be broad-based to reflect its participatory, demand-driven market orientation. The decree settingupthe ARCN needsto bereviewedto reflectthe proposed change in paradigm. The new ARCN should have a legal framework that allows it to generate funds aggressively from diverse sources. As sole manager of all public funds for research and development, ARCN should allocate some funds competitivelyand some throughcontracts. The focus of research needs to be extended to include sectors other than crops (e.g., livestock, fisheries) as well as post-harvest activities (e.g., processing, marketing, agribusiness). Appropriately targeting and focusing research will require strategic priority setting in terms of commodities, regions, productive activities, and types of technology. Process changes will also be necessaryto bringabout a reversalfrom the current centralized, top-downapproachto priority setting, execution, and evaluation of research to a more inclusive and participatory bottom-up approach. 29 Reforming extension Focus of action should include improvingcoordinationand reducing duplicationof effort in the ADPs, improvingthe financial sustainabilityof extension services, increasingthe accountability of extension agents to farmers and agribusiness firms, and diversifying the national extension strategy away from its focus on crops by providingservicesthat meet a broader range of needsof farmers and agribusiness firms. Improving coordination. To avoid duplication of effort, the mandates of all public agencies, private firms, and civil society organizations involved in agricultural extension activities at the local level should be reviewed, and, if necessary, redefined. The federal government should move more aggressively to implement the recommendations of the Agricultural and Rural Institutions Review to improve the delivery of extension services by eliminating the ADP monopoly, diversifying sources of service provision, and evolving new roles for the ADPs through participatoryapproaches. For extension services to reach rural households, appropriate policies must be put in place to strengthen the linkages between national, state, and local agencies. The federal government should reinforce its policies for decentralizing service delivery to the states, LGAs, and communities. Producerorganizations, too, shouldbe invitedto play an active role inextension. Improving jnancial sustainability of agricultural extension services. Public agricultural extension services are not financially sustainable in their current form. The federal and state governments have been unable to fully fund the ADPs, and extension services deliveredthrough the ADPs are steadilybeingeliminated, especiallythose providedto the poorestproducers. Ifthe current trend is to be reversed, government bodies with responsibility for promotingtechnology transfer will have to adopt policies and mechanisms for pluralistic funding. In the case of commercial farming, approaches involving user fees should be explored. In certain cases, privatization of extension services delivery should be considered as a potential vehicle for increasing efficiency, accountability, outreach, and competition. Although privatization may appear an extreme option, it is importantto note that most agriculturalinformation services are providedby privateagents. Public agencies should concentrateon providingtechnicalsupport to privateproviders, developingpublic-privatepartnerships for service delivery, sharing information with privateproviders, andestablishing co-financingmechanisms. Holding extension agents accountable to clients. Introducingcompetitioninto service provision will help make extension agents more accountable to farmers and agribusiness firms. Competition should come from multiple sources, including the public, private, and nongovernmentalsectors. Experience from many other countries as well as from within Nigeria shows that extensionservicescan bedelivered effectivelythroughmany non-traditionalchannels. Diversifying the national extensionstrategv Inextension, as inresearch, the emphasis needsto be expanded to respond to a much broader range of needs of farmers and agribusiness firms. This implies that the major extension service providers must develop new skills relating to: management of soil and water resources, management of livestock and fisheries enterprises, management of post-harvest value-adding activities (e.g., processing, storage, preservation, packaging), and management of marketing and distributionactivities. The nature of extension service provision will also have to change. Extensionagents will require training in adaptive participatory research methods, which will assume increasing importance in tomorrow's more client-orientedtechnology generationandtransfer systems. 30 Strengthening Input Supply Systems Seed and Planting material. Efforts to strengthen seed distributionsystems should be led by the privatesector, butnewlyestablishedprivateseed companies face many obstacles andmayrequire public support duringan initial phase. This supportmust come intwo forms: (i) establishmentof a conducivepolicyenvironment; and (ii) investments in selectedareas to strengthensupply public as well as demand. To provide a conducive policy environment, in additionto improvements inthe general business climate, government needs to enact transparent and non-burdensome regulations relating to varietal release and registration, seed certification and quality control, and seed health monitoring.Duringthe early "emergence stage" of seed industry development, public investment will be needed at two stages of the seed multiplication process: (a) Plant breeding research. Since the Nigerianseed market is still very small, privateseed companies are unlikelyto recover the full cost of research through seed sales. Public investment in plant breeding research is, therefore, needed to produce improved germplasm; (b) Source seed production. Productionof commercial seed entails multiplication of several generations of source seed (breeder seed, foundation seed, seed of hybrid parental lines). Public investment in source seed production is neededto generatethe seedusedby privatecompanies as parent seed. Increasingthe supply of improvedseed and plantingmaterials will not be successfulunless there is effective demand. Since many farmers in Nigeria still lack knowledge of improvedvarieties, an effective promotion campaign will be needed to publicize their potential benefits. Public research and extensionorganizations can contributeto this campaignby conductingparticipatory on-farmtrials, plantingdemonstrationplotsfeaturingnew varieties, and distributingseed samples along with technical information about crop management practices. Over time, as farmers become more familiar with improvedseedand demandbeginsto strengthen, opportunitiesshould be exploredto publicize and promote new varieties throughpublic-privatepartnerships involving public extensionworkers andprivate-sectorvarieties. Fertilizer An urgent priority should be to stimulate the development of an efficient and profitable fertilizer industry. This will requirea two-prongedapproachdesignedto strengthen demand for fertilizer on the one hand and improve the supply of fertilizer on the other. At the same time, fertilizer interventionsshouldbe developedwithin the context of a wider strategy that recognizes the importance of complementary inputs, output market development, and an appropriate sequencing of interventions(for an account of lessons learned from efforts to promote fertilizer use inAfrica, see Box2). Farmers' effectivedemand is the ultimatedriving force in all input supply systems, and financial profitability of fertilizer use by farmers is the foundation of a sustainable fertilizer promotion strategy. Since effectivedemandfor fertilizer is currentlyvery low inNigeria, actions undertaken to improve supply will be unsuccessful unless they are accompanied by activities that can increase effective demand at the farm level. The public and private sector both have a role to play in strengthening demand, although this role will vary throughtime. During early stages of fertilizer market development, state-levelpublic research organizations should take the lead in evaluating alternative fertilizer management strategies and developing fertilizer use recommendations tailored to local needs and circumstances. State- and local-level public extension services should communicate these recommendationsto farmers and demonstrate their potentialprofitability. Over time, as farmers become more skilled in the use of fertilizer, they come to appreciate the benefits of fertilizer, and begin to purchase fertilizer regularly, the 31 generation and transfer of technical information related to fertilizer use can be assumed by the same private firms that sell fertilizer to farmers. These efforts to strengthen demand for fertilizer must be accompanied by parallel efforts to improve supply. Fertilizer, like seed, is a private good, so there is little rationale for public involvement inthe fertilizer sector. Fertilizer market developmentefforts should, therefore, have the clear aim of creating viable private-led production and distribution systems. As with seed, however, public support will be neededduring an initial phaseto allow emerging private fertilizer companies to become established and expand their operations to a commercially viable scale. This support will likely have to come in two forms: (i)establishment of a conducive policy environment; and (ii) public investmentsin selected areasto strengthensupply as well as demand. Protection of inefficient local manufacturing plants is not justified, and the government has rightly discontinued production of urea at the NAFCON facility and has sharply scaled back production of phosphatic fertilizers at the FSFC facility, while exploring options for privatizing both facilities. Inthe near term, the lowest cost option for procuring fertilizer will continue to be to import. Over time, the feasibility of reviving some level of domestic manufacturing capacity should not be ignored, especially if international prices of petroleum and petroleum derivatives remain strong. Domestic manufacturing, however, is unlikely to become attractive until economies of scale can be achieved through expansion of domestic and possibly regional markets. Government also needs to scale back its involvement in distribution. The current system under which public enterprises distribute fertilizer-often at subsidized prices4iscourages the development of competitive private fertilizer marketing companies, distorts the price signals needed to guide efficient allocation of a scarce and costly resource, and creates numerous opportunitiesfor rent-seekingand corruption. At the same time, given the current underdevelopedstate of the Nigerian fertilizer market and the weak level of demand, it is clear that the private sector will be unable to fill the void created by the exit of public enterprises. Successful development of a viable private fertilizer industry will therefore require structured dissolution of state enterprises in combination with policy reforms and investments designedto create incentives and raise capacity for the private sector to step in. Capacity building can be pursued through targeted investments, such as the donor-supported DAIMINA project, which provides resources, methods, and tools in support of business development activities. Inmany parts of Nigeria and for many crops, fertilizer use will not be profitable unless progress can be made in reducing the high internal handling and transportation costs that leave farmers paying prices that can be doublethe landedimport cost or even more. Expandingirrigation capacity Given Nigeria's abundant land and water resources, there is room for considerable expansion of irrigation capacity. Experience shows, however, thatjust because irrigation is technically feasible does not mean that it is economically profitable or institutionally sustainable. Inundertakingthe development of Nigeria's irrigation potential, it will be important to ensure that investments in irrigation are appropriately targeted in terms of location and scale. In addition, policies and procedures must be put into place to ensure that water is managed efficiently and equitably. Finally, a comprehensivenational water policy must be enacted that can set overall priorities for irrigation development, coordinate the activities of the many players who are active in the sub- 32 sector, ensure a proper enabling environment, and address the provision o f public goods and services neededfor the successful development of the sub-sector. Scale of irrigation. Expansion of irrigation capacity should take careful note of economies and diseconomies related to scale. Investment in new infrastructure for medium- and large-scale irrigation should be undertaken only after careful analysis has shown that the investment i s justified from an economic point of view. Participation in existing large public schemes is often financially attractive to farmers because they are not being charged the full cost of water, but most public schemes are not viable from an economic point of view because of the high levels of capital investment. Over the past 15 years, the development of small-scale, farmer-owned and operated irrigation systems has been quite successful. Located mainly in fadama areas, small-scale irrigation systems are based on low-cost pump technology, and they allow more flexibility in water management than do the medium- and large-scale public irrigation schemes. A distinguishing feature that has contributed to the success of small-scale schemes i s that they take into account the needs of the local communities, who gain greater control over water management decisions. In the short run, the main strategy for expansion of national irrigation capacity should be accelerated expansion o f the area under small-scale systems, Key actions should include: (i) support for private advisory services development to meet producers' needs; (ii)finance rural infrastructure to link production points and marketing centers; (iii) support for rural financial and insurance services development22; (iv) set up a clear property rights framework and conflict- resolution mechanisms between the different fadama user groups, especially farmers and pastoralists; and (v) monitor of water table levels and ground water quality. New Infrastructure rehabilitation vs rehabilitation of existing inj?astructure. With medium- and large-scale irrigation, the rehabilitation of existing schemes can offer more attractive prospects than investment innew schemes (FAO, 2000b). Selection of schemes for rehabilitation should be based mainly on technical and economic aspects, although rehabilitation should focus on modernizationto account for present and future needs, rather than the needs for which the system was originally designed. Also, considering the political sensitivities relating to water access and use as well as the multiple and competing interests regarding its allocation, investments in irrigation needto be linkedalso to broader issues o f development and political economy. The Review of Public Irrigation Sector inNigeria (ROPISIN) has recommended the rehabilitation of 12 schemes covering 52,600 ha. This would nearly double the area under active irrigation by medium- and large-scale schemes. 22Less than 2 percent o f rural households have access to formal banking services, and insurance services are virtually non-existent. 33 Box 2: PromotingEfficient and Sustainable Fertilizer Use -Lessons from Africa Crop yields in Nigeria are low and stagnating partly because o f soil fertility problems, which can be addressed by promoting greater use o f fertilizer. However, fertilizer use is unlikely to increase unless efforts are made to address the many underlying structural problems affecting fertilizer demand and supply. A recent World Bank review of fertilizer promotion schemes in Africa identified 10 guiding principles that can be used to design and implement public interventions to support efficient and sustainable growth infertilizer use. 1. Wider sector strategy: Interventions designed to promote increased use o f fertilizer should be developed within the context o f a wider sector strategy that recognizes the importance o f complementary inputs, output market development, and an appropriate sequencing o f interventions. 2. Market friendly: Interventions designed to promote increased fertilizer use should be designed to support market development and not undermine incentives for private investment. 3. Competitive: While the private sector should be inthe driver's seat, markets should be competitive to ensurethe lowest cost and best quality service. 4. Demand driven: Farmers' effective demand, shaped by the financial profitability o f fertilizer use, should be the ultimate driving force o f input supply systems and the foundation o f a sustainable fertilizer promotion strategy. 5. Economically efficient: Interventions to promote the use o f fertilizer should be carried out only where it is economically efficient to use fertilizer. 6. Farmer empowerment: Interventions designed to promote increased use o f fertilizer should empower farmers to make their own decisions on the most appropriate way to improve soil fertility in their particular context. 7. Exit strategy: Public sector interventions designed to promote increased use o f fertilizer should be designed with a clear exit strategy, except for a few long-run public good functions, such as, market regulation, infiastructure development, and R&Dfor natural resource management. 8. Regional integration: Countries should seek regional integration and harmonization o f fertilizer policies to reap economies o f size, which are especially important ina region such as Afiica with many small countries. 9. Sustainable: Interventions to promote increased use o f fertilizer should be economically, institutionally, and environmentally sustainable. 10. Pro-poor: Assuming the previous nine guiding principles have been followed, another important consideration is that public interventions designedto promote increased use o f fertilizer should also aim to promote pro-poor growth. In exceptional circumstances, poverty reduction andor food security objectives may be given precedence over efficiency and sustainability goals, if it can be determined that fertilizer interventions are the most cost-effective way o f addressing these problems. ~ ~ Rehabilitationofexistingmedium- and large-scale schemes shouldbe undertakenkeepingin mindthe followingprinciples: a) Schemesshouldbe consideredfor rehabilitationonthe basis of an analysis oftheir economicviability andfinancial attractivenessfor farmers. b) Schemes should be simple, user friendly, and conducive to the participatory irrigation management(PIM) approachadoptedby Government. c) Scheme rehabilitationwould requiremodernizingexisting infrastructureand redesigning to allow lower cost gravity irrigationand farmer management as well as to reduce the energy costs ofpumpedschemes. 34 d) Rehabilitationldesign of any public scheme should include improvements of rural infrastructure, particularly, rural roads and market infrastructure, and would require the strengtheningof advisory andrural financial services. Some schemes may prove unsuitablefor rehabilitation and should be allowed to revert to rainfed production. For schemes which are basedon largedams impounding large quantities of water but having only little infrastructure, a number of options may be considered: (i)pumping from reservoirs with small farmer-owned pumps to shoreline plots; (ii)controlled water releases downstream to allow for micro-scale extraction, flooding of fadama areas (for flood recession production and recharge of aquifers, (iii)construction of small downstream weirs where applicable applicable for gravity irrigation; (iv) rehabilitationhedesign outlet structures to allow for gravity irrigation to the extent possible; and (v) promotion of fishproduction. Capacity Building. Potential productivity gains associated with irrigation will be realizedonly if water can be used efficiently. This requires that management be efficient at two levels: at the system level and at the farm level. At the system level, water user associations offer the most promising strategy for empowering groups of farmers to manage irrigation efficiently so as to reduce costs, improve irrigation services, and raise productivity. The optimal size and allocation of responsibilities to such associations depends on local conditions and capacities, but several approaches have been piloted successfully in Nigeria and provide good models for scaling up. This will require training designed for Water Users Associations covering all aspects of establishmentand operation of schemes. At the farm level, increasedwater use efficiency can be encouragedby rational pricing policies that require farmers to pay the full cost of water. Appropriate water pricing policies must be supported by use of improved management practices that minimize unnecessary water losses, such as, appropriate cropping patterns, cultivation methods, and soil moisture conservation practices. As irrigation expands, many farmers will require training in irrigation techniques, water scheduling, estimation of crop water needs, and maintenance of tertiary and secondary canals and structures. Comprehensive water resources policy and strategy. Sustainable development of Nigeria's irrigation potential will require a conducive policy environment and institutional changes. In addition, it is unrealistic to expect that the irrigation sub-sector will develop if it must continueto depend on irregular, unpredictable, and often inadequate public budget allocations. With assistance from FAO, the FederalMinistry of Water Resources is formulating a national irrigation policy and strategy, which is needed by FMWR to enable it undertake its mandate to efficiently manage the nation's water resources. Among other objectives, the new policy would serve to resolve the issue of competing demands among domestic users, help resolve international water rights issues in the Lake Chad and NigerBenue Basins, address environmental shortfalls, remove inter-state rivalries over water allocations, and ensure that adequate attention is given to hydropower generationand inland waterway transportationneeds. Underthe new national irrigation policy, it is anticipatedthat government will shift from beingan operator to being an enabler, with a much more focused role of federal public agencies as bulk water service providers. A major goal of the new policy would be to shift the focus of the River BasinDevelopment Authorities (RBDAs) to concentrate on the macro and public goods aspects of water management, while continuing their activities on the management of larger public schemes. The new focus would include: (i)developing river basin plans; (ii) coordinatingwater resource management across river basins; (iii)monitoring water quality and quantity; (iv) operating reservoirs; and (v) supervising dam safety. The RBDAs would devolve operation of schemes to farmers' organizationsand privatecompanies as local capacitiesbecome stronger. 35 TECHNICAL ANNEX 3-A: DEVELOPING EFFECTIVE ROAD MAINTENANCE SYSTEMSINTANZANIA Like most African countries, Tanzania has waged a long battle, still unfinished, to develop adequate systems for maintainingthe roadnetworkneededto serve a large and widely distributed population. The World Bank's 1994 WDR recommended a dual approachto resolvingthe road maintenance problem- with equal emphasisto reliability in the flow of funds and efficiency in their use - and citedTanzania as a best-practicecase. Within a year or two, however, it became clear that institutionalcapacities inthe country were insufficient to bringthe new systems started to effective fruition at the same time as implementing a large roads investment program, and serious questions arose about major corruptionand misuse of funds inthe Ministry of Works. A new approach was gradually worked out, not fundamentally changing the directions which had been adopted in the early 1990s but filling them out and giving them a much stronger constitutional foundation. December 1998 saw Parliamentary passage of the Road Tolls (Amendment) No. 2 Act, formally establishing a Road Fund - in lieu of the one created by official declarations ofthe FinanceMinistry in 1991-92 - and a Boardof nine persons (including an independent chairperson appointed by the President, four representatives from the private sector and four senior civil servants) to run it. And on July 1 2000 TANROADS (Tanzania National RoadsAgency) was established as a semi-autonomous agency ofthe Ministry of Works (as permittedby the ExecutiveAgencies Act of 1997) to manage maintenance and development ofthe primaryroadnetwork. The mutually supportive public- and private-sectorefforts engendered by these arrangements have yielded several promising trends. The Road Fund Board's extensive monitoring and auditing efforts help significantly to ensure that user charges to support maintenance are duly collectedand devotedto the intended purpose; besides regularfinancial audits of recipients (and itself), a major technical audit was contracted with a Norwegian/SouthAfrican consortium in 2002/03, reachingthe conclusionthat the Boardwas generally receivingvalue for money andthat 90-95% of activities supported were achieving the required standard of output. Annual expenditures for road maintenance, which had risen with the original Road Fund from totally inadequate $10 million or less at the start of the 1990s to $30 million equivalent in 1994/95 and then fallen off with the problems of that period, reached $48 million in 2000/01 and $57 million in 2002/03, benefiting in part from Finance Ministry acceptance of the increase in fuel levy recommended by the Board for the latter year. Whereas road works had previously been undertaken almost entirely by foreign contractors or by force account, from the early 1990s considerable effort had been devoted to supporting development of local contractors, especially for maintenance. Their confidence, and possibilities, suffered severe blows from the unexpected drop-off in funding after 1995, but TANROADS has resumed training efforts and, in line with general government policies, about 90% of all road maintenance efforts are contracted out. Combined effort of TANROADS and the Board has yielded a significant reduction in vehicle overloading(fines for which accrue to the RoadFund) interms of both scale and incidence, such that only 6% of the one million vehicles weighed in 2003/04 were found to be overloaded, but coverage still needs to be extended to further roads. The long-discussed Road Maintenance Management System, based on HDM 4, was finally established in one zone in 2002/03, and extended to the other three the following year, and the corresponding data collection has been activelyunderway, with an updatednetworkinventorycompletedinDecember2003 andwork on traffic counts andpavement quality startedover the lastyear. Trends in overall networkconditionunfortunatelycannot be establishedwith any high degree of reliability, even for the primary network, because of uncertainties as to its actual extent, the predominanceof gravel and earth surfaces subject to rapid change with weather conditions, and 36 the unavoidable element of subjectivity injudgments. The most valid information available is probably that on the actual length (as opposed to percentages of the varying network assumed at different times) that was rated `good' or `fair' in regional engineers' visual inspections. These numbers show relatively little change in the extent of primary network rated `fair' (from about 9,000 kms in 1990 to 10,700 kms in 2004) but much sharper growth and fluctuation inthe extent rated `good'. It increased from less than 3,000 kms in 1990 to more than 7,000 kms in 1993 before falling over the second half o f the 1990s and rising again thereafter to reach more than 11,000 kms in 2004. While there i s some room for doubt about the validity o f the last figure (since it i s much greater than the previous year's estimate even though gravel road maintenance had fallen much short of plan in the interim), it indicates that good primary network may have increased as much as 8,000 kms since 1990 (which can be compared, for order o f magnitude, with the increase of 5,000 kms in the same category in the similarly sized country of Ethiopia between 1995 and 2002). Among the further developments that are beingconsidered, the most important may be to ensure the most efficient possible allocation of the scarce resources available for maintenance. A very important initiative by the Board has beento start systematic financial support for maintenance of local roads by allocating 30% o f Road Fundresources to district and urban councils, as required under the law. The Board considers that these allocations are sufficient for maintenance of only about one-quarter o f the approximately 50,000 k m s of such local roads, while those for TANROADS can cover about half o f the 28,000 kms for which it is responsible. The full inventory of local roads presentlyplanned will clearly be very important for identifyingrational priorities in use o f the expanding resourcesmaintenance badly needs- and inthe development of local capacities for maintenance management which are still very weak. TANROADS' Maintenance Management Systemwill be of great value to this endtoo. The 2003 technical audit o f Board expenditures also urged greater use, in the Performance Agreements which the Board negotiates each year with the agencies that manage the maintenance, of clearly defined and objectively measurable indicators of accomplishment - and greater readiness to withhold or reduce disbursements when monitoring reports show lack of progress. While TANROADS' effectiveness has clearly benefited from its more autonomous status, and the improved salary structure and stronger discipline that this made possible, consideration is also being given to the possible advantages of further increasing its independence and flexibility for delivering against a firm set of overall performance targets agreedwith its supervisingMinistryandthe Board. 37 TECHNICAL ANNEX 4-A: SUMMARY OF ENTERPRISE-RELATED TAXES The basic corporate profitstax rate inNigeria is 30 per cent. The company income tax is paid by companies incorporated inNigeria on all its profits wherever they have arisen and whether or not they have been brought into or received in Nigeria. For a foreign company, the following profit will be deemed to be derived from Nigeria: (a) where company has fixed base of business inNigeria, profit that is attributable to that fixed base; (b) where it does not have such fixed base in Nigeria but operates through agent inNigeria, profit that is attributable to business carried on by such agent; (c) where its business in Nigeria involves single contract for surveys, deliveries, installation or construction, profit from the contract; and (d) where activities are between company and another person controlled by it or which has controlling interest in it and Board is o f opinion that business has not been conducted on arm's lengthbasis, profit as adjusted by Board to reflect arm's length basis. Companies may also pay tax on turnover where true profits of company cannot be readily ascertained - usually 20 percent o f company's turnover will be deemed to be its profits (the law stipulates that a reasonable percentage o f turnover be used to determine an acceptable profit). Value-Added Tax (VAT): The VAT (5% although under reform plans, it is anticipated that this rate will be increased to lo%, except for basic food, medical and pharmaceutical products, baby products, books and educational services and personal rent) was introduced in Nigeria in 1993 and was intended to overcome the problems in the use of the sales tax system. The rationale behind replacing Sales Tax with the Value Added Tax was informed by a number o f factors, including - (i)the base o f the Sales Tax operated under Decree No.: 7 of 1986 was narrow an covered only nine categories of goods plus sales and services in registered hotels, motels and similar establishments. VAT provides a broader base and includes most professionalservices and banking transactions which are high profit-generating sectors. (ii)only locally manufactured goods were targeted by the Sales tax Decree of 1986, but under VAT, a considerable part of the tax to be realized was from imported goods; (iii) VAT is based on the general consumption since behavior of the people, the expected high yield from it would boost the revenues to the state governments with minimumresistance from the payers of the tax. The VAT system in Nigeria is administered by the Federal Inland Revenue Services (FIRS). The VAT is administered centrally by the Federal Government usingthe existing tax machinery of FIRS in close co-operation with the NigeriaCustoms Service andthe State InternalRevenue Services23.The net proceedsfrom the VAT accrue solely to the State Governments after makingan allowance of 15 percent to cover the Federal Government costs of administration. Nigeria's tax code includes a capital gains tax of 10 percent, and a withholding tax of 10 percent on dividends and rents withheld at source24. Capital gains tax is assessed on the disposal of financial and material assets, includingoptions, debts, property, and foreign currency. Rental income, including for the lease of transportation and machinery, i s taxed at 10 percent. Nigeria imposes a royalty withholding tax of 15 percent. An education tax o f 2 percent is levied on company profits in Nigeria, and several business transactions are subject to the payment of stamp duties, which vary in the amount and formula for calculation. Excise taxes of 40 percent apply to tobacco and alcohol products. 23The sharing formula for the VAT is 15-50-35 for Federal, State and Local government tiers respectively. 24This is not a separate tax but a mere advance payment of tax, normally offset at the time of payment of final tax assessed, usingthe withholding tax credit note issuedat the time of withholding. 38 State taxes - states collect a variety of taxes including capital gains tax; stamp duties; and a businessregistrationlevy. Each state can definewhat qualifies as an urbanandruralarea, and the tax varies based on where a business is located. In urban areas, the annual business registration fee will be a maximum of N10,OOO for the initial registration and N5,000 for each renewal thereafter. In rural areas, the fee is N2,000 for the first registration and N1,OOO thereafter. Local governments, including villages, also collect local taxes. This tier of government has the right to collect at least 18 different types of taxes and fees including shop and kiosk registration; right of occupancypermit; signboardand advertisingpermits 39 Table 2: ~ ~ ~ ~ aofr PartsEntrynChargesat Lagos and Qtker West Africail Ports ~ o Port Index of Port Chargesper 17O# TEU Abidjan = YO0 Abidian 100 B, Ensc af Trading Figure 1: Iridicators oil EaseofTrading, ........................................................... 0 20 .............................................................................................. 40 60 80 .............................................. 100 ' Source: Doingbusiriess 7006 C. Import ClearanceSequencefor Non-EPZ Rank Activity Time Cost Lapse 1 pro-forma invoice sent to clearing agent 2 clearing agent obtains and completes form M Clearing Agent purchases cargo insurance and submits Pre-shipment 3 receipt to bank Procedures 4 Bankapproves Form-Mand sends to other agencies PSI agent assigns cargo an SEIC number and completes 5 inspection and issues CRI Clearing Agent obtains single goods declaration form and with documents providedcalculates duties owed for company to pay at a bank 7 Clearing agent takes all documents to CPS at Port 8 Deputy comptroller reviews documents 9 Desk Officer verifies documents Accounting Officer verifies documents and gives to Post shipment 10 Deputy comptroller for final verification procedures 11 ASYCUDA assigns serial number 12 Manifest department certifies Bill o f Lading 13 Customs examination shed officers check cargo against documents 14 Valuation Unit Inspects Cargo and release 15 Clearing Agent confirms payment and claims goods Source:FIAS 2000 As seen from figure above, the sequence of activitiesassociatedwith import-exportclearance for non-EPZ firms is complex. Some 71 signatures on 13 different documents are required for imports into Nigeria. Clearing and forwarding agents note that importers typically avoid the principal Import document, Form M, and its financial obligations by dealing with port-based syndicates. For a fee of N1,OOO - N4,500 these syndicates provide a Form M and forge bank signatures and stamps. Procedural requirements are also overlappingat times. For example, after a clearing agent takes the completed documentation to the Deputy Comptroller for review, the package is passed round to both a desk officer and the accounting office before being returned again to the deputy comptroller for final verification. It is only then that they are passed to an officer at ASYCUDA. 41 TECHNICAL ANNEX 4-C: GENERAL AND SECTOR SPECIFIC LICENSING BURDENS 1.GeneralLicensingRequirements License Government level Frequency LandRent S A VAT F A Corporatetax F A educationtax F A CapitalGains F A Nigeriansocial insuranceTF F M IndustrialTraining Fund F M Vehicle Registration F A License for logo advertisementon vehicle L A Vehicle pollution inspection F A Hackneypermit for privatebusinessvehicle F A Canteen license L A Property tax S A Electricity charges 0 M Water Charges 0 M Factory/planthealthcertificate L A Factory/plantenvironmentfee S M LandRent L A Combinedexpat residencypermit F A CERF'AC - reentrypermit F A Expat quotafee F A Propertytax F A NEPZA registratiodlicense 0,F A CapitalAllowance F A Surcharge on import duty F E Corporatetax returns F A Fire and safety license S A PAYE S A CorporateWithholding tax F E Source: Value Chain 42 LicenseburdenandFees for Shrimp Industry Government License level Frequency Official Fee FactoryPlant health certificate NPA A 3500 Jetty License NPA A 0.35m Wharfage license NPA E 5000 Pilotage license NPA A 20590/1 Lighterterminal levy NPA A Documentation fee NPA E 500 Vessel survey certificate F A 500001cerl Berthingpermit F E 4000 vessel sailing permit F E 2000 Vessel departure permit NPA E Vessel arrival permit NPA E Pilotage fee NPA A 2059011 Cabotage License F A 2% turnover BunkeringApproval Fee F E 0.35 per litre Radio license F A 4800 Petroleum License F A 10000 Vehicle Entry License Fee F A 5000 Source: Value Chain PermitRequirementsfor TouristEstablishments No. Tax 1 Veterinary Fees 2 Registration for Issuance o f HealtWSanitation 3 Business Premises 4 Sign Board and Advert Permit 5 Tenement Rate 6 Environmental Protection board, Sewage and Refbe Disposal 7 NTDC Registration Fees 8 VATITAX 9 Industrial Training Fund 10 Pay as you earn (PAYE) 11 National Social Insurance Trust Fund(NSITF) 12 company's Income Tax 13 On and OffLiquor License 14 Radio and Television License Fees 15 Education Tax 16 NEPA Bills and Diesel 43 TECHNICAL ANNEX 4-D: COMMERCIAL AND INVESTMENT LEGALFRAMEWORK LegalFrameworkfor Contract Enforcementand Settlement of Disputes The 1999 Constitutiono f the Federal Republic established and conferred specific jurisdictions on the following courts: (a) the Supreme Court of Nigeria; (b) the Court of Appeal; (c) the Federal High Court; (d) the High Court of the Federal Capital Territory Abuja; (e) a High Court of a State; (f)the Sharia Court of Appeal of the Federal Capital Territory, Abuja; (g) a Sharia Court of Appeal of a State; (h) the Customary Court of Appeal o fthe Federal Capital Territory, Abuja; and (i)a Customary Court of Appeal of a State. In addition to the courts established by the Constitution, there exist Magistrate, Area and Customary Courts which are established by the National Assembly in respect of the Federal Capital Territory, Abuja and the State House of Assembly inrespect o fa State. These courts have limitedcivil and criminaljurisdiction conferred on them by their respective enabling law. Magistrate Courts in Lagos for instance have their jurisdiction to hear civil matters that relate to contracts or commercial activity limitedto claims of not more than W250,OOO. There are also a number of specialized courts such as: (a) the National Industrial Court (NIC) established by Trade Disputes Act which plays a significant role inthe mitigation and settlement of trade and labor disputes; (b) the Investment and Securities Tribunal (IST) established under the Investment and Securities Act (ISA)to provide an accessible, cost-effective and timely as well as efficient means of redressing all civil matters arising out o f investment and securities transactions: (c) the Code o f Conduct Tribunal established by the Constitution to try public officers who are in violation o f the Code of Conduct; and (d) any other Tribunal that may be constituted by an Act o f the National Assembly or the Law o f the House o f Assembly of a State. CommercialDisputeResolution: Nigerian courts have full jurisdiction over commercial matters with the exception of matters involving the revenue of the Federal Government and or its agencies, taxation o f companies, admiralty, customs, companies affairs, banking regulation, intellectual property, citizenship and immigration, bankruptcy and insolvency, aviation, drug and poisons, weights and measures, mines and minerals, including oil and gas and any action involving the Federal Government or any of its agencies, which are subject to thejurisdiction of the Federal HighCourt. Alternative Dispute Resolution: In addition to the traditional court system, there exists also formal Alternate Dispute Resolution (ADR) mechanisms established by law or under the authority of a law. These include include: 0 The Arbitration and ConciliationAct: The ACA provides for commercial arbitration only in respect o f which the parties must have agreed in writing to arbitrate. The ACA provides for Domestic, International, Ad hoc and Institutional Arbitration. It was enacted to implement Nigeria's treaty obligations under the Convention on the Recognition and Enforcement of ForeignArbitralAwards made inNewYork on the 1 0 of June 1958 (the ~ New York Convention). The legislation was to apply to international commercial arbitrations and incorporate the basic concepts of the United Nations Convention on International Trade Law (UNCITRAL) Model Law on International Commercial Arbitration of 1985. 44 0 Multi-Door Court Houses: Inan effort to reduce the number of cases filed inthe court system and reduce the backlog, Lagos and Abuja States have each introduced a court annexed Multi-Door Court House (MDC) as a form of alternate dispute resolution system. The MDC providesa one-stop menu option of arbitration, mediationand early neutral evaluation.All MDC settlements are endorsed by the ADR judge and registered in the MDC registry as a consent order, which is enforceable by the court. The MDC charges an admin fee plusADR service fee. Citizens Mediation Centers: Another form of ADR is the Citizens Mediation Centers (CMCs) that are currently established and operating in Lagos State by the Ministry of Justice andbeingintroducedto other States.The CMCsprovidemediationservicesto the general public free of charge on mixture of business and social matters including landlordtenant issues, monetary claims, family issues, and employer/employee issues. The terms of the mediationis recorded in a Memorandumof Underrating(MOU) which could be registered in with the court registrar and is enforceable by law. The CMCs is achievingoutstanding success with thousands of cases beingmediated annually, and the average time it takes to finalize the settlement ranges between 30 minutes in simple matters to maximumoftwo monthsin other matters25. Customary Practices: The law courts recognize customary law which is consensual rather than adversarial in approach. The High Courts will enforce settlement under customary law, although this would be more effectively done if traditional leaders recordedthe settlements suchthat they couldbeusedas evidence. The PrincipalLaws GoverningEnterpriseActivity: Law of Contract: There is presently not a single law in Nigeria, specifically governing the creation, performanceand discharge of contracts.NigerianContract law consists of the received English law, doctrine of equity, applicable English laws, a number of Nigerian legislation and court decisions. Nonetheless,there are instanceswhere due to the peculiarnature of the Nigerian economy and culture, some differences do occur.26 Sale of Goods: The law applicable to sale of goods inNigeriais the Sale o f GoodsAct, 1893 of the United Kingdom. The law applies to sale of goods only and not to contracts for work and materialswhich is governedby the generalrules of contract. A number of states havepassedtheir own Sale of Goods laws, butthese are mainly reproductions ofthe Sale of GoodsAct, 1893. Hire Purchase Act: The Hire Purchase Act seeks to protect the hirer in such classes of hire purchasetransactionsto which it applies. Companies and Allies Matters Act: All business enterprises governed primarily by the Companies and Allied Matters Act (CAMA) and must be registered with the Corporate Affairs Commission. Available forms of business organizations include: (a) Limited Liability Companies; (b) Unlimited Liability Companies; (c) Companies Limited by Guarantee; (d) Partnerships; (e) Sole Proprietorships; (0 IncorporatedTrustees; (8) Branch or Subsidiaries of ForeignCompanies; (h) RepresentativeOffices. ''Martin Pierce, ChukwuemekaEz, Ngozi Okoh, Evaluation ofthe Lagos State Citizens' Mediation Centre (16Ihto 25Ih February 2006; A study commissionedby DFIDBritish Counsel. 26Kehinde Aina, Nigeria: Legal Barriers to Contract Enforcement, April 2006. Report prepared for the World Bank. This report is quoted extensively throughout this section. 45 Investment and Securities Act: This is the principal legislation governing corporate amalgamations and reconstructionsincluding internal reorganization, liquidation, mergers, and takeovers. The objective of the Act is to provide a more efficient and viable capital market positionedto meetthe country'seconomic anddevelopmental needs. Foreign Investment: Non-Nigeriansare permittedto invest and participate in the operation of any enterprise in Nigeria. Where, however, a foreign company wishing to set up business operations in Nigeriamust be incorporatedfirst as a Nigerian branch or subsidiary. The locally incorporated branch or subsidiary company must then apply to the Nigerian Investment PromotionCommission(NIPC) for BusinessPermitandother requisitepermitsand licenses. Nigerian Investment Promotion CommissionAct: Created to be a one stop investment agency, NIPC was established amongst other things to co-ordinate, monitor, encourage and provide necessaryassistance and guidancefor the establishment and operation of enterprises inNigeria.It i s also mandated to act as a liaison between the Federal Government and the general business sector to formulate policy matters, including fiscal measures which would promote the industrializationofNigeriaor the generaldevelopmentofthe economy. Foreign Exchange (Monitoring and Miscellaneous Provisions) Act: This Act established the Autonomous Foreign Exchange Market (AFEM). The Act permits the investment of foreign currency or capital imported into the country by an individualor corporate entity provided the investment is imported through an `Authorized Dealer" (a bank or non banking institution so licensedby the CentralBank ofNigeria).The Act also permitsa person, notwithstandingwhether or not he or she is a resident or citizen of Nigeria, to deal in, invest, acquire or dispose of, create or transfer any interest or other money market instrument whether dominated in foreign currencies or otherwise. Trustees Investment Act: This Act sets out the type of investment to which possible investors may make in public investible funds in government securities and securities of public quoted companies. This is particularly important where a foreign investor wishes to invest in debt securities, unittrust schemes or investment fundoperation. Intellectual Property Rights: The applicable laws governing the protection of intellectual propertyrightspertain:(a) The PatentsandDesigns; (b) Trade Marks; and(c) Copyright. National Office of Technology Acquisition and Promotion Act: Every contract or agreement entered into by any person in Nigeriawith another person outside Nigeriainvolvingthe transfer of foreign technology to Nigerian partners shall be registered with the National Office of Technology Acquisition and Promotion(NOTAP) in the prescribed manner not later than sixty (60) days fromthe execution or conclusion ofthe agreement. Insolvency and Bankruptcy: Differentlaws apply to insolvent individuals and corporate bodies. The Bankruptcy Act is generally applicable to individuals whilst the Companies and Allied Matters Act applies to corporate bodies. Other relevant laws include: (a) FailedBanks (Recovery of Debts) and FinancialMalpractices in Banks; (b) Mortgage InstitutionsAct; and (c) Nigerian Deposit Insurance CorporationAct. Pension Reform Act: This Act repealedall previous legislationsregulatingthe administrationof pension benefits in Nigeria, its primary objective is the encouragement of savings among employees so that in retirementthey are not impoverishedandthe establishment of a uniformset 46 of rules, regulations and standards in the public and private sectors o f the Nigerian economy on matters of pensions. Workmen's CompensationAct: This Act provides for the payment o f compensation to workmen for injuries suffered in the course of their employment. It makes provision on a graduated scale basedon the extent of injury suffered. Informal AlternativeDisputeResolutionMechanisms Eightyto ninety percent ofNigerians rely on customary law. The law courts recognize customary law which is consensual rather than adversarial in approach. The High Courts will enforce settlement under customary law, although this would be more effectively done if traditional leaders recorded the settlements such that they could be used as evidence. There are also a multitude of non-formal dispute resolution systems that are available to the enterprise sector. These can vary across states and industries. These systems include the Chartered Institute of Arbitrators (UK) Nigeria Branch; Lagos Regional Center for International Commercial Arbitration; Negotiations and Conflict management Group (NCMG) Center for Dispute Resolution; Market Associations; Trade Unions; Nongovernmental Organizations (NGOs); Cooperatives; Local Government Officers; Traditional Leaders; and Nigerian Customary Arbitration. Arbitration is conducted under the Arbitration and Conciliationact of 1988,which requiresthat Arbitration clauses be made standard in all contracts. Awards are bindingexcept for point of law or due to arbitration negligence. One major disadvantage with ADR i s the lack of enforcement, except via the courts which in turn undermines the "out o f court" advantages that arbitration seeks to provide. There are also a multitude of non-formal dispute resolution systems that are available to the enterprise sector. These can vary across states and industries. The table below summarizes some o f the more recognized systems inoperation inNigeria at thistime: 47 Characteristics of non-formal dispute resolution mechanismsaccessibleby MSME's Dispute Principal advantages Principal DisputeWithin Dispute with resolution disadvantages Community OutsideParty mechanism Market associations informal, quick, focuses on tends to be adjudicative in Chairmanand Council Chairmenfrom each interests of traders and nature, skills of Chairmen electedby market traders. Association meet market, respected intuitiverather than trained Hear parties and decide and agreeoutcome case in interest of parties ininterestof all and other traders in same markets. Ifnot market. successful, parties may go to police but risk BDRand bribery on both sides. 'Never' go to court. Trades Unions informal, quick, focuses on tends to be adjudicative in Chairman of Union Ifaccessible, interests of traders nature, skills of Chairmen decides as per Market Chairmenof each intuitive rather thantrained Association Unionmeet and decide as per Market Association NGO informal, quick, effective Outreach can be limited where inhandsof well- and dependenton donor trained mediators support Cooperative informal, everyonewell Meet parties all together Will try to find known to each other, andwork out a mutually personin authority quick, supportive for acceptablesolutionfor the over other party and benefitof whole group meettogetherto community resolveproblem Local Government known inthe community Canprovideopportunity Although has no authority May meetwith Omcers for corruption to adjudicate, is respected LGAs from other and may want to help by area ifphysically suggestingsolutionto all possible parties. Traditional leaders quick, enforceable, Leadersskills often Meets partieswith Joins with respected intuitive rather than Council, hearsand decides TraditionalLeader trained. May not be aware case, imposes penalty if from other ofparties' legal and human appropriate. Decision community to rights. enforcedby decideoutcome. If community\ccourts. not appropriate, partieswill haveto go to court, 48 TECHNICALANNEX 4-E: SOME BESTPRACTICECASE STUDIESIN INVESTMENTCLIMATE 1.OnTradeFacilitation: A. Tunisia: In the late nineties, cargo spent an average o f 8 days in Tunisian ports- and in many cases up to 18 days, compared with a few hours in Singapore. Moreover, in Tunisia, as inNigeria, 7 different ministries were involved in inspections, using a predominantly paper system encouraging corruption and inefficiency. What did Tunisia do? The project took gradual approach to allow for demonstration effect to counter complaints - mostly from customs agency who feared loss o f control. Introduced reform concentrated first on administrative procedures- single automated form for all agencies. Looking at customs, Tunisia managed to get the time to import down from 8 to 3 days in five years. The manifest processing time also decreasedfrom 2 days in2000 to just 2 days in2005. Inthe area o ftechnical control, Tunisian cargo was subject to 50-80% inspections in2000; now this figure is 10.3%. Riskmanagement techniques introduced. B. Kenya: Kenya Revenue Authority instigated the Revenue Administration and Modernization Program-as part o f this they launched the customs Reform and M o d project to 1) improve revenue collection and service delivery; 2) reduce lead time in cargo clearance 3) minimize human intervention at the points o f entry and exit; 4) ensure transparency and predictability o f customs processes. This was achieved by restructuring the authority, simplifying procedures, installing an electronic data interface, scanners for inspections, post-clearance audit etc. The program, SIMBA consists o f 4 parts; tradex, orbus, paybox, leuk- doc posting, networking, electronic payments and reference guide. Progress so far has included; a new customs Services Department and the first part o f SIMBA, TRADEX introduced July 2005. Training centers have also been established in strategic places have beenestablished andthere are already results; o Reduction intime taken to assemble and process documents from 62-56 (4 day drop) o Reportedreduction inoccasion for extra-legal behavior as less face to face opportunity o Reported increase inrevenue from X to Y. ''Exceptcustoms procedures 49 2. On ImprovedAdministrativePractices Examples of Public-Private Sector Collaborationin Support of Improved Business AdministrativeProcesses 1. It is often important to focus on mechanismswhich "ring-fence" private sector activities from a potentially predatorygovernment or from systemic corruption. *:* Two recent examples come from Cambodia. The first was an initiative of the Garment Manufacturers Association of Cambodia, GMAC, who gained approval from the Minister of Commerceto station GMAC representatives within the Ministry wearing red armbands, to monitor the processing o f export licenses, and to reporton any corrupt practicesto the Minister. 9 The secondwas aneffectiveprivatizationof labor inspectionsingarmentfactories, also inCambodia, with a tripartite governance structure overseeing the work of ILO monitors who replaced inspections from the Ministry of Labor (and, increasingly now, from overseas buyers), using a transparent monitoring system and reducing opportunities for shakedown. The key to success behind both these examples seems to have beenenhancedtransparency, reducingopportunitiesfor graft. 2. In developingan Agenda of priority reforms, active cooperation with and "inclusion" of private sector can be crucial to ensure the ownership and disseminationnecessary to ensure sustainable performance. Selecting the optimal vehicle for this to happen can be the difference between success and failure as there are no "cookie cutter" solutions and each country, state, and industry will have customized design requirements. For instance, prior to reform, Kenyan businesses had to face over 1,300 licenses and fees, implementedby 178 different state bodies. It was decided to establish a public-private task force set up and a review of current practice was completedwithin eight weeks. The task force usedthe "regulatory guillotine approach" to identify and eliminate those procedureswhich were no longer servingany usefulpurpose. Source: Reportedby FIAS, April 2005 and AFTPS 3. On Tax Reform Use of the Marginal EffectiveTax Rate (METR) can be an effective way o f measuring the tax burden and on establishing benchmarks for reform. This was estimated inthe case o f Senegaland over a two year period from 2003 to 2005 the authorities were able to reduce the METR from 48% down to 28% with further reductions now targeted through additional reductions in the corporate tax level from 30 down to 25%. The application o f METR targets in Nigeria, accompanied by programs o f transparency and outreach where details o f the taxes payable are clearly available to businesses, offers a real opportunity to increase business confidence and grow the tax basethrough increased formalization. The METR approach- Tax reform in Senegal Akin to the current situationin Nigeria, investment in Senegal was until recently being deterred by its tax system, characterizedby multiple and high tax rates, heavy administrativeratesandrent-seekingbehavior. In responseto this however, Senegal launchedatax reforminitiative based on indepth diagnostic work undertaken incollaborationwith FIAS2000, establishingaworking group dedicatedto improving the tax systemfor enterprise growth. Their aim, as supported by the WBG through the Private Sector Adjustment Credit in 2004, was to implement a private enterprise tax regime that will stimulate investment. Specificallythe objective was to reduce the marginal effective tax rate (METR) to enterprises in Senegal from 45 to 28 percent, an optimal figure determined by FIAS analysis. The METR measures the fiscal fee on an investment decision for industry when no investment incentivesare used. The measurementcaptures the impactof tax on the marginal investment or the last unit of capital invested. This includes not only the effects of several taxes and fiscal treatments on productive investment, but also on economic andfinancialfactors. Widely supported by both the Senegalese Government and the private sector, the corporate tax rate was legally reducedto 28% February 2004, with recognitionof increasedtransparency inthe system. This should improvenot only the playing field for enterprises, but international opinion as reflected in the improvement expected in the Doingbusinesstables for 2007. In this vein, Nigeria could measure the baseline and try to incorporate specific incentives into the tax code, decrease top rates and reduce the number of taxes, resulting in a reduction of the METR to a point that will help stimulateor addressthe needs of certainsectors and privateactivities. 50 TECHNICAL ANNEX 5-A: BANKING SECTOREFFICIENCY This section briefly describes the evolution of the Nigerian banking system from 1985 to 2005, thus providing the context for CBN's decision to increase the minimum capital requirement to NN25 billion, which then triggered consolidation. It goes on to use historical data provided by CBN to analyze efficiency in intermediation. Specifically, we examine the determinants of interest spreads via accounting decompositions and regression models to get a sense of how efficiently the banks that resulted from the consolidation process will intermediate funds to borrowers. The section closes with an analysis of whether the way in which new banks were formed (e.g., via capital injection from shareholders versus merger with multiple partners) has affected marketperceptions, and thus has implications for post-consolidationperformance. A. Background Inthe contextof a 1986 Structural Adjustment Program,Nigeria undertookbroadfinancial sector liberalization, including easing entry restrictions into banking. Low entry requirementsand high potential rentsfrom arbitrage activities inthe foreign exchange market encouragedthe creation of new banks, andthe total number ofbanks grew from 40 in 1985to 119by 1991(Lewis and Stein, 2002). In addition to the incentives provided by the foreign exchange market, bouts of macroeconomicinstability also made lending (other than short-term) less attractive, and the level of private credit drifted down from about 20% of GDP in 1986 to near 10% in the early 1990s (Figure 1). The Nigerian banking system suffered from high fragmentationand low intermediation.The large number of banks also taxed supervisory capabilities. Inflation began to climb in 1990, reaching 70% in 1995, and private credit levels fell below 10%of GDP. The combination of fragmentation and macroeconomic instability eventually contributedto systemic distress. Caprio and Klingebiel (2003) classify Nigeria as experiencing a systemic banking crisis, which they define as much or all of banking capital being exhausted, throughout the 1990s. In 1995, at the peak of the inflationary spike, those authors report that almost half of the banks reported being in financial distress. Between 1994 and 1998, half of the banking system sector faced problems of insolvency. Though the number of banks declined in the wake of the crisis, it remained over 80, more than twice the level prior to structural adjustment (Figure 1). Since the crisis, annual inflation has remained low by Nigerian standards, averaging near ten percent, and private credit levels have crept up, albeit slowly. However, in the first years of the new millennia, the Nigerian banking sector was still characterizedby high fragmentation and low levels in intermediation. 51 Figure 1: Evolution of the Nigerian Banking Sector n 80.0% 140 70.0% 120 60.0% 100 g v) 5 50.0% z 0, 40.0% 80 u- 5 0 -+Inflation (CPI) 2..30.0% 60 --tBanks .-g E 20.0% 40 1 CI 5 - 10.0% 20 E I 0.0% 0 1985 1990 1995 2000 Against this backdrop, the Governor of the Central Bank ofNigeria(CBN) announcedon July 6, 2004 that bankswould be requiredto achieve minimumcapital ofNN25 billion by December3 1, 2005. The requirement had been NN2 billion. The increase was intended to help bringabout a diversified, stable financial sector that would ensure the safety of deposits, while at the same contributingmoreto economic development via intermediation.The bankswere expectedto: "shore up their capital through the injection of fresh funds where applicable, but were most importantly encouraged to enter into merger/acquisitionarrangements with other relativelysmaller banksthus taking advantageof economiesof scale to reducethe cost of doing business and enhance their competitiveness locally and internationally." (CBN Press Release, January 3, 2006). The announcement met opposition from some corners, including the banks themselves.28The Chartered Institute of Bankers of Nigeriaissued a positionpaper requesting an extension of the deadline to achieve the new minimumcapital requirementfrom 2005 to 2006, and seekingto re- categorize banks into Investment, Universal, and Mega banks, each with its own capital requirement. Some in the Senate also introduced measures that would have compelled CBN to change its stance. However, CBN held firm, and in the end the number of banks shrank from 89 to 25 through mergers, acquisitions, and other capital injections involving 76 banks which together accountedfor 93-5%of total deposits. Therefore, despite all of the potentialdifficulties, a relativelysuccessful outcome to the consolidationprocess was achieved within the timeframe originally stipulatedby CBN. B.EfficiencyofIntermediation,2000-2005 Cross-countryevidence indicates stronglythat highinterest spreadslimitthe breadthand depth of bankingsectors (Figure 2). In particular, countries with higher spreads have substantiallylower levels of credit to the private sector (as a share of GDP). In this section, we use bank-leveldata from CBN to calculate interest spreads for Nigerian banks from 2000 to 2005. We calculate the spread as the differencebetween interest income received on loans (divided by total loans) and the interest expense paidon deposits (dividedby totaldeposits). Though spreadshave declined in 28See NigeriaBusinesshfo.com, "N25 Billion Naira Capitalization: The Journey So Far and its Likely Implicationson theNigerian Economy." July 11,2005. 52 recent years (Table l), remain high in comparison with those for the countries in Figure 3, they thus explaining, inpart, Nigeria's low level o fprivate credit relative to GDP. 0 40 - 30 - 0 20 - 0 8 10 - 0 0 0 0 0 -00 0 - I I I I This graph is taken from Beck and Fuchs (2004). It plots Private Credit/GDP and interest spreads for a sample o f 83 countries for 2001. Data are from International Financial Statistics and the Financial Structure Database update from the World Bank. We also offer a simple decomposition o f those spreads to better understand the factors driving both their high level and the decline (Table 1). Because the implicit lending rate that we calculate relies upon interest income earned, it is an ex-post measure o f both the interest rates charged and loan repayment rates. In a decomposition exercise, it is preferable to use the ex-ante interest rates charged, and then include loan loss provisions among the factors that could explain the spread.29 We lack bank-level data on ex-ante lending rates, and thus we assume that the implicit rate we calculate from ex-post interest income realized already accounts for portfolio quality. We do not therefore account for loan loss provisions in the decomposition. W e do however account for overhead costs, taxes, and required reserves. The overhead costs are those attributable to loans, which we identify by calculating the share o f loan interest revenue intotal revenue. Profit margin is a residual after adjusting for the tax rate, reserve requirements, and overhead^.^' Deposit rates in Table 1 decline throughout the period, but lending rates drop more sharply, resulting in the decline in spreads. That decline can not, however, be easily explained by declining overhead costs, required reserves, or taxes, all o f which remained roughly at the same level. The decline in profit margins, therefore, has been driven by reductions in lending rates. A 29 See, for example, Beck andFuchs (2004). 30 The formulawe use is: Profitmargin=(1-tax rate) x (implicit lendingrate-implicit deposit rate/(1-ReserveRequirement)-operating costs/loans) The tax rate is calculatedfrom actualtax payments. The reserverequirement is 10%. Operatingcostsare those attributableto lending, andthus are equalto the share of income from lendingmultipliedby totalcosts.Notethat the figures inTable 1come from simple averagesacrossbanks. Thus, while the profit marginis an accountingresidualfor each individualbank, it neednotbe equalto a simple residualfor the sector as a whole. 53 part of those declines is no doubt attributable to a prolonged period of relative macroeconomic stability. Another part is likely due to increasedcompetition amongbanks. Table 1: Interest Spread Decomposition 2001 2002 2003 Implicit Lending Rate 32.14% 36.59% 27.24% Implicit Deposit Rate 12.00% 13.43% II.37% Spread 20.14% 23.16% 15.87% Overhead Costs 6.59% 8.38% 6.96% Taxes 0.16% 0.09% -0.11% Reserve Requirement 1.33% 1.49% 1.26% Profit Margin 11.29% 12.30% 7.20% 0bservations 317 341 34 1 Source: Authors' calculations from CBN data. We use quarterly observations Within the sector, there are substantial differences inthe cost structures of large and small banks that are reflected in lower borrowing and lending rates for large banks (Table 2). Banks in the largest size quartile pay substantially less for their deposits (5.6%) than do smaller banks (9- 10%). Their implied lending rate is also somewhat lower than that of smaller banks, and thustheir spreads are low relative to most of the other banks. A part of the difference in spreads can be explainedby the low overhead costs of the large banks. Those overheads are so low, in fact, that the largestbanks enjoy higher profit marginsthanthose inthe other size quartiles. Table 2: InterestSpreadDecomposition,2005 Q3, By Assets I Smallest Largest lst-25th I 26th-50th I51th-75th 176th-IOOtP percentile percentile percentile percentile Implicit Lending Rate 25.11% 21.73% 25.61% 19.48% Implicit Deposit Rate 8.69% 10.19% 10.20% 5.60% Spread 16.42% 11.54% 15.41% 13.88% Overhead Costs 9.46% 7.14% 8.83% 4.72% Taxes 0.21% -0.35% -0.07% 0.09% Reserve Requirement 0.97% 1.I3% 1.I3% 0.62% Profit Margin 6.80% 2.84% 5.42% 7.56% 0 bservations 21 21 21 21 Source: Authors' calculations from CBN data The decompositionresults inTable 2 are suggestive of the benefitsof having larger banks, due to their lower lending rates, spreads, and overheadcosts. However, because those lending rates are based on ex-post interest income, we would prefer an analysisthat explicitly controls for portfolio quality, We therefore offer regressions that control for multiple bank characteristics, including loan loss provisions. Those regressionsalso include macroeconomicvariables that were likely to affect the spreads chargedby all banks. The basic regression model is based on that used in Martinez Peria and Mody (2004) for developingcountries inLatin America:31 Spread ,, = a /?,overheads + ,1+ /?,liquidity ,,+ P3equity,,+ /?, provisions ,, /?,market share ,, + /?,erest income +/?,loan share +/?,Herfindah1 +P9Tbill rate +PI,inflation,, int + /?,,industrial production + E,, 3 1That modelis motivatedby the dealershipmodelof banks spreadsdevelopedinHo andSaunders (1981), inwhich banks are risk-aversedealerstrying to balanceloananddeposit markets.Because loanrequests anddepositflows can be asynchronous,spreadsare seen as fees chargedby banks for the provisionof liquidity under uncertainty.See MartinezPeriaandMody (2004) for further descriptionof the modeland extensions by other authors. 54 Where the interest spread is calculatedas described above for bank i at time t. Overheads are the ratio of overhead costs to total assets. As in the simple decomposition, we expect that higher overheads costs are passed onto borrowers inthe form of higher spreads. Liquidity is the ratio of liquid assets (cash and depositswith other banks) to deposits. Inthe Latin American context, high liquidity was thought to inflict a cost on banks, since a bank must forego the opportunity to hold a higher-yielding instrument. Thus, Martinez Peria and Mody (2004) hypothesize that banks will try to transfer this cost to borrowers, resulting in a positive association between liquidity and spreads. Similarly, those authors hypothesize that there is an opportunity cost associated with holding excessivecapital, andthus they expect a positive relation betweenequity and spreads. IntheNigeriancontext, thekeyregulatorychangewasthe increased capitalrequirement,which a number of banks exceeded by a substantial margin. In satisfying that requirement, many banks also became flush with liquidity, but they were not bound by regulation to maintain liquidity at those levels. In addition, many of those banks experienced substantialdeposit inflows during the consolidation period, which also boosted liquidity, as depositors fled to banks they perceivedto be of higher quality. A number of banks were, therefore, looking for outlets for excess liquidity, putting competitive pressure on lending rates. We hypothesizethat liquidity would be negatively associated with spreads inNigeria during this period. Banks that were in this situation were also likelyto have high capital shares, andthus we also expect equity (bank capital plus reserves, over total assets) to be negatively associatedwith spreads. Provisions are the ratio of loan lossprovisionsto total loans, our measureof portfolio quality. We expect that banks with higher loan loss provisions would have lower interest income from lending, andthus lower spreads, since we calculate spreads on an ex-post basis. By controlling for the part of the interest spread attributableto portfolio quality with this variable, we obtain more reliable estimates of the effects of other variables (most notably overhead costs) on the spreads that were chargedex-antethanwe could inthe simple decompositionexercise. Market share is the bank's share of total banking sector deposits, our measure of bank size. To the extent that larger banks can take advantage of economies of scale, we would expect market share to be negatively related to spreads. This expectation is also consistent with the decomposition in Table 2. Martinez Peria and Mody (2004) note, however, that market share could also be equated with marketpower, andthus the ability to charge higher rates on the loans. The coefficient on this variable will therefore indicate which of these two hypotheses is better supportedby the data. We control for bank orientation using interest income, the ratio of total interest income to operating income, and intermediation, the ratio of net loans and advances to total liabilities. Using bank-level data across countries, Laeven and Levine (2005) demonstrate that specialized loan-makingbanks have different performancecharacteristicsthan specialized investmentbanks, and that loan-making banks tend to have a higher share of interest income. Because our spreads are calculated from interest income on loans, we expect competitive pressure in the lending market to be better reflected inthe banks specializedin that area, and thus we expect a negative association between interest income and spreads. Similarly, we expect those banks that lend a relatively high share of their available liabilities to be most responsive to the same competitive pressures, and thus a negativerelation betweenintermediation and spreads. Following the literature, in some specifications we include a control for banking sector structure and three macroeconomic control variables. Herfindahl is a standard index of sector 55 concentration, which we calculated based on bank shares of total deposit^.^' If deposits are concentratedin the hands of a few banks, those banks might be able to drive up lendingrates, as they control the supply of funds. We would therefore expect a positive relation between concentration and spreads. The macroeconomic controls are the T-bill rate, inflation, and industrialproduction. The T-bill rate is the rate of interest on short-termtreasury bills, which is includedas a proxy for the marginalcost of funds faced by banks. lnflution is includedbecause price shocksmightnotbepassedthroughequallyto the nominallendingandborrowingrates, and thus these differentialeffects would be reflectedinthe spread. Industrialproduction is an index measuringthe change inindustrialproduction,which is includedto capture businesscycleeffects that are reflectedin spreads.As an economy slumps andproductionslows, borrowers become less creditworthy, andthus banks must charge higher lendingrates, which are then reflectedinhigher spreads (all else equal). The Resultsappear inTable 3. All models are estimatedvia OLS, and includebank-specificfixed effects. The estimated coefficientstherefore reflect departures from each bank's average spread for the period. The models also include year and quarter dummies to capture trends and shocks that affected all banks. Coefficientson the year dummies tell a story broadly consistent with the decomposition in Table 1, with spreads beinghighest in 2001-2002, and lower at the end of the period. In the first two models we use the `narrow' spread calculated as in Martinez Peria and Mody (2004), which i s basedon interest income only from lending, and interest paidon deposits. Loan loss provisions are negative and highly significant, as hypothesized. That coefficient should therefore control for the downwardpressure that poor portfolio quality puts on spreads that are calculatedusingex-post data. The positive, highly significantcoefficient on overheads, therefore, provides more direct evidence that cost reduction is associated with lower spreads than did the decomposition. The negative, significant coefficient on liquidity supports the hypothesis that banks that had a high share of liquid assets tended to charge lower spreads (relative to their average for the period), perhaps because of the competitive pressures associated with expanding their lending. The coefficient on equity is also negative, though it does not achieve significance in any of the specifications presented here. These results stand in stark contrast to those found for Latin America, where the signs for liquidity and equity are positive, and liquidity is highly significant. We attribute this difference, in part, to the unique incentives facing banks as part of the consolidationexercise. The coefficient for the market share of total deposits is negative and significant in model 1, though insignificant in model 2 when the macroeconomic controls are introduced.The negative coefficient is consistent with large banks enjoyingeconomies of scale or scope that reduce their spreads, apart from those already reflectedin their overhead costs. Finally, and as hypothesized, banksthat earn a highershare of incomevia interestandthat lenda higher share oftheir liabilities charge significantly lower spreads. These results are consistent with the idea that banks specializingin lendingare facing more competitivepressurethan other banks, which is reflected intheir spreads. 32The index is calculated by summingthe squaredmarket shares ofall banks. 56 Table 3: Spread Regressions larrow interest larrow interest Vide interest Vide interest Zxplanatory Variable pread pread pread pread 3verhead costs/ assets 1.121 1.509 0.619 0.883 (0.009)*** (O.OOO)*** (0.004)*** (O.OOO)*** iquid assets/ deposits -0.013 -0.012 -0.005 -0.005 (0.022)** (0.042)** (O.OOO)*** (O.OOO)*** -oan Loss Provisions/ Loans -0.075 -0.072 -0.029 -0.029 (O.OOO)*** (O.OOO)*** (O.OOl)*** (0.001)*** Market Share of Deposits -0.011 -0.009 0.000 0.0005 (0.027)** (0.196) (0,991) (0.819) zquity -0.065 -0.075 -0.026 -0.027 :capital+ reserves/ assets) (0.308) (0.344) (0.391) (0.411) nterest income share -0.006 -0.007 -0.004 -0.004 (0.055)* (0.038)** (0.117) (0.089)' ntermediation -0.466 -0.475 -0.062 -0.07 :loans/ liabilities) (O.OOO)*** (O.OOO)*** (0.003)*** (0.001)*** ierfindahl for Deposits -0.0001 0.00001 (0.391) (0.743) ?ea1Treasury Bill Rate 0.003 0.002 (0.108) (0.006)*** nflation 0.003 0.002 (0.242) (0.014)** 2hange in Industrial Production -0.001 -0.00001 (0.493) (0.980) 2001 0.029 0.007 0.021 0.015 (0.065)* (0.781) (O.OOO)*** (0.090)* 2002 0.054 0.023 0.026 0.019 (0.004)*** (0.414) (O.OOO)*** (0.048)** 2003 -0.027 -0.044 -0.003 -0.003 (0.122) (0.140) (0.650) (0.800) 2004 -0.015 -0.031 -0.007 -0.008 (0.443) (0.373) (0.313) (0.512) 2005 -0.027 -0.013 -0.023 -0.006 (0.186) (0.743) (O.OOl)**' (0.619) Cluarter 1 0.018 0.009 0.005 -0.001 (0.068)' (0.475) (0.138) (0.830) 3uarter 2 0.000 -0.001 -0.006 -0.009 (0,990) (0.959) (0.220) (0.060)* Cluarter 3 0.009 0.012 -0.001 -0.003 (0.268) (0.155) (0.759) (0.411) Zonstant 0.342 0.33 0.108 0.065 (O.OOO)*** (O.OOO)*** (0.000)'"" (0.027)** 3bservations 1838 1779 1838 1778 Vurnber of banks 91 91 91 91 ?-squared 0.208 0.222 0.212 0.248 ?obust D values in parentheses * signifidant at 10%; **significant at 5%; *** significant at 1% All modelsestimatedvia OLS, with bank-specific fixed effects We recognize that Nigerian banks have historically earned a large share o f their income from sources other than lending. To get a broader sense o f the efficiency o f intermediation, we therefore calculate a 'wide spread', which includes interest income from all earning assets and interest paid to all interest bearing l i a b i l i t i e ~Results are qualitatively very similar to those for . ~ ~ the narrow spread, except that the wide spread coefficients are smaller (in absolute value). This could be because our independent variables were chosen to explain lending spreads, and thus 33Specifically, total earning assets includes total due from other banks, short-term investments, certificates of deposit, discounted bills, other financial instruments, net loans and advances, and investments(other than short-term). Total interest bearing liabilities includes money at call, interbanktakings, total deposits and takings, total due from other banks, certificatesof deposit, other loans and advances from financial institutions, debentures, and other liabilities. 57 their effects are dampened when we examine a broader range o f income-generating activities. Two ofthe macroeconomic controls(inflationandthe Treasury bill rate) are also significant, and of the hypothesizedsigns inthe wide marginregressions, whereas they hadbeeninsignificant for narrowmargins. Inall, the similarity betweennarrowandwide marginresults inspires additional confidenceinthe basic structure ofour To the extent that the historicaldata provide a reliable guide to future performance, the models presented here suggest that the consolidationexercise could make financial intermediationmore efficient by reducing spreads. In particular, the larger banks, which tend to have lower overheads, also tend to charge lower spreads. Many of those banks also found themselves with highliquidity as a result of consolidation.Consistentwith the view that those banksare searching for lendingand investment outlets for those resources, liquidity is strongly negatively related to spreads. Amongthe group ofhighly liquidbankswith low overheads, those that earn a highshare of income from interest and lend a high share of their total liabilities are also likely to face the most downwardpressureon spreads.As notedinthe CEMtext, while narrowingspreads are seen as positive from the perspective of prospective borrowers, they do put pressure on the banks themselves. As profit margins narrow, bank distress becomesmore likely, and the increased size of these banks means that a single failure poses greater systemic risk than it might have in the past. C.MarketReactions to Consolidation To this point, we have stratified banks by size for convenience, and in part because bank identitiesare maskedinthe dataset providedby CBN. We recognize, however, that inadditionto size, the structure of post-consolidationbanks has affectedmarket perceptions. To illustrate, we draw upon data from a second source, Bankscope, which identifies banks by name. We can therefore mergehistoricalfinancial informationfrom banksthat were mergedas of December 31, 2005, to get a sense of the consolidated banks' future profitability and activities, and market reactionsto their share offerings. It is clear that the consolidationhas produced a number of different structure types within the 25 banks which ended with more than NN25bnof capital. The absolute amount o f capital is one of the few features whichthe bankshave incommon. The following structures are identifiable: Group 1: The largesttraditionalbankshaveachievedthe capitalthreshold moreor less on their own. They havetaken the opportunityto consolidate long-established affiliates and perhaps acquire one or two much smaller banks. Within this group there is one significant alliance between one of the "old" established banks and one of the newer banks (Standard Trust and UBA). In such cases, management remains stable and controlledfrom an identifiable"lead" bank. Group 2: The banks in group 1 have beenjoined at the top of the system by a small number of newer banks which have taken the opportunity of consolidationto make an exponential jump in their capitalization and potential. There is clear unity of managementdirection. The rest of the banks have achievedthe capitalthreshold by forming groups. Here there may well be problems bringing together the various banks into a coherent whole and there may be disturbances as the different management groups vie for influence. Some 34Inunreported specifications,we usedathird, wider spreadthat includedincomefromfees and commissions,and derived qualitatively similar results. 58 smaller groups are clearly made up of voluntary partners, and the market perceives that they have a reasonable chance of success (Group 3). The market is less happy about the prospects for largergroupings which may have been broughttogether mainly by force of circumstance (Group 4). In the figures that appear below, existing banks that were formed from four or more banks are placed in Group 4, those from three or fewer in Group 3. Finally, a small number of foreign owned institutionshave decided to stay independent and wait to see if a further opportunitypresents itself(Group 5). Table 4 assigns banksto the respectivegroups. Table 4: Bank Groups Group 1: FirstBank,Guaranty Trust, UBA,Union Group2: Afiibank, Intercontinental,Oceanic, Zenith Group3: Access, Diamond,Ecobank,ETB,Fidelity,IBTC-Chartered,Platinum-Habib,Wema Group4: FCMB, First Inland, Skye, Spring, Sterling,Unity Group 5: NIB/Citi, Stanbic, StandardChartered Detailedpre-consolidationfinancial informationon eachofthe banksthat becamea partofone of the 25 banks listed in Table 4 is available in Appendix 1. Here information is summarized by bank group. For each current bank that was formed via merger, we calculate its financial indicators as the weighted average of the indicators of its component banks. The weights are based on the pre-consolidation assets of the component banks. The figures that appear in the tables beloware simple averages acrosseachbank group. The market seems to be less happy about the prospects for larger groupings of banks. Table 5 shows that banks in Group 1were 3-4 times as large as the other domestic banks in Groups 2-4, in terms of bothpre-consolidationassets and equity. The foreign banks (Group 5) were smallest, at roughly half the size of the domestic banks outside Group 1. While the value of domestic banks' share issues from 2003 to 2005 was similar across bank groups, the rate of subscription varied. Issues from banks in Group 1 and 2 were oversubscribed by a wide margin. The was only 69.4% -- supporting the view that the market has less confidence in the larger subscriptionrate for those in Group 3 was also reasonably strong (95.6%). For Group 4, the rate groupings. Table 5 B a n k G r o u p V a l u e o f % P r e - P r e - c o n s o l S h a r e Issues Subscribed C o n s o l i d . E q u i t y 2 0 0 3 - 5 ( a v g ) ( a v g ) Assets ( a v g ) (avg) G r o u p 1 2 0 0 0 0 1 4 0 . 9 % 3 0 8 2 3 4 3 1 2 0 2 G r o u p 2 2 0 3 5 2 1 5 3 . 3 % 1 2 0 8 8 9 1 1 7 5 1 G r o u p 3 1 5 6 3 8 9 5 . 6 % 6 8 0 9 4 8 8 5 7 G r o u p 4 2 0 5 0 4 6 9 . 4 % 7 4 9 3 8 9 1 1 4 G r o u p 5 n.a. n.a. 4 2 6 0 8 5 1 3 7 Pre-consolidationprofitability figures provide further support for the market's relatively tepid reaction to the share offerings from larger groupings of banks (Table 6). Group 4's return on equity was only 20.8%, compared with 36.7% for Group 2, and 31.5 % for Groups 1 and 5. 59 Similarly, Group 4's average NPL share was 19.9%, higher than for any other bank group. In many respects, however, banks in Group 3 have a profile nearly identical to those in Group 4. In fact, their average return on equity (19.4%) was lower than that of Group 4 banks. Perhaps the higher subscriptionrate for the Group 3 share issues is due to having fewer merger partners, and thus fewer perceivedpost-mergermanagementdifficulties. Table 6 B a n k G r o u p R O E 2 0 0 4 , R O A 2 0 0 4 , 70 NPL o r l a s t o r l a s t 2 0 0 4 , o r l a s t G r o u p 1 3 1 . 5 % 3 . 1 % 1 7 . 5 % G r o u p 2 3 6 . 7 % 3 . 5 % 1 4 . 7 % G r o u p 3 1 9 . 4 % 3 . O % 1 8 . 2 % G r o u p 4 2 0 . 8 % 2 . 7 % 1 9 . 9 % G r o u p 5 3 1 . 6 % 3 . 9 % 1 1 . 7 % Notes: ROE, ROA, and NPL figures are calculated from Bankscope data It is also worth noting that the banks made up of smaller and larger groupings, that is, those in groups 3 and 4, intermediate funds differently than other banks (Table 7). While deposits represent 65-71% of assets across all domestic bank groups (1-4), banks inGroups 3 and 4 lend a higher share of those funds. Loans are 41-44% of assets for Groups 3 and 4, only 32% for banks in Groups 1 and 2. Deposits in other banks comprise a high share of the assets of foreign banks (Group 5, 43%) and Group 1 banks (28%). Our understandingis that many of those deposits are used to finance letters of credit overseas. In that sense, the figures in Table 7 indicate that the banks in Groups 3 and 4 are the most involved in domestic intermediation. Instability within those groups could imply some disruption for borrowing firms and individuals. Table 7 B a n k G r o u p L o a n s / D e p o s i t s / D e p o s i t s i n Assets 2 0 0 4 , A s s e t s 2 0 0 4 , 0 ther o r l a s t o r l a s t B a n k s / A sse ts ( 2 0 0 4 ) , o r l a s t G r o u p 1 3 1 . 8 % 6 5 . 2 % 2 8 . 1 % G r o u p 2 3 1 . 7 % 7 1 . 4 % 1 3 . 2 % G r o u p 3 4 1 . 2 % 6 7 . 5 % 1 9 . 7 % G r o u p 4 4 3 . 6 % 6 6 . 9 % 1 6 . 5 % G r o u p 5 2 9 . 1 % 4 5 . 1 % 4 2 . 7 % A clear focus for on-going analysis of the process of consolidation is to see whether the different structures yield different results for their owners: and in particular whether the prejudicesof the market are supportedby the data. 60 References Beck, Thorsten, Robert Cull, and Afeikhena Jerome. (2005). "Bank Privatization and Performance: Empirical Evidence from Nigeria." Journal of Banking and Finance, 29: 2355-2379. Beck, Thorsten and Michael Fuchs. "Structural Issues in the Kenyan Financial System: Improving Competition and Access." World Bank Policy Research Working Paper 3363, July 2004. Caprio, Gerard and Daniela Klingebiel, 2003. "Episodes of Systemic and Borderline Financial Crises." World Bank, mimeo. Databaseavailable at: http://econ.worldbank.org/ external/ default/ main?theSitePK=47806O&contentMDK=2 - 0355185&menuPK=546095&pagePK=64168182&piPK=64168060 Ho, Thomas and Anthony Saunders. (1981). "The Determinants of Bank Interest Margins: Theory and Empirical Evidence." Journal of Financial and QuantitativeAnalysis, 4: 581- 600. Laeven, Luc and Ross Levine (2005). "Is There a Diversification Discount in Financial Conglomerates?" Journal of Financial Economics,forthcoming. Lewis, Peter; Stein, Howard (2002): "The Political Economy of Financial Liberalization in Nigeria," in: Stein, Howard; Ajakaiye, Lewis, Peter (Eds.): Deregulation and Banking Crisis in Nigeria, Palgrave. Martinez Peria, Maria Soledad and Ashoka Mody. (2004). "How Foreign Participation and Market Concentration Impact Bank Spreads: Evidence from Latin America." Journal of Money, Credit, and Banking, 36(3): 511-537. 61 .. 3 X TECHNICALANNEX 5-B: DEVELOPINGTHE MARKET FORMORTGAGE FINANCEINNIGERIA Nigeria has a large housing supply deficit in rural and urban areas. Estimates of the current national housingdeficit range between 12 - 16 million housingunits. Addressing the housing needswill require significant financial resources and it is evident that public funds alone will be insufficient to provide adequate financing to respond to the demand for housing in Nigeria. Promoting access to affordable housing finance through the private sector will be key in achievingthe Government's goal to increase the availabilityof adequatehousing. The authorities have made the facilitation of privatesector housingsupply and housingfinance a key element of the 2004 NationalHousingPolicy. The policy document identifies the need to (a) promote of capital markets and financial products to provide long-term funding, (b) reform Government sponsoredagencies in the housing sector, and (c) revise the laws and regulationsgoverningland title andtitle transfer as priority areas. Increasingthe capacity of local financial markets to finance home ownership will foster social cohesion. Many Nigerianhouseholdshave stable and secure incomes and could affordadequate mortgages, if loan-amortization were spread over a long enough horizon to reduce mortgage paymentsto serviceable levels. Local banks estimatedthat their target market for mortgageloans i s 10to 20 million customers. As in other developingcountries, a large share of savings in the Nigerianeconomy is bound in home equity. Inthe absence of adequate long-termfinancing, investments in housingstock take place on a "build-as-you-earn" basis. The lack of functioningmortgage markets also constrains the ability of home owners to leverage their savings in home equity to increase investment. Developing the mortgage market is an important tool in the support of small and medium enterprises, as the mobilizationof home equity often provides the start-up capital for small and mediumentrepreneurs. Inthemediumterm, especially ifcurrenthighinterestratesprevail,theaffordabilityofmortgages will be restrictedto higldmiddle-incomegroups. Thus there is a needto develop instrumentsand policiesto support increasedaccess to mortgage finance for lower income groups. As elaborated below, the Government can both help commercial banks expandtheir client base `down-market' and supportmoregrass root financial institutionsindevelopinghousingfinance productsto serve lower-incomegroups. Market structure and capacity To date the provisionof loans for housinghasbeennegligible. Total mortgageloans extendedby private or public financial institutions are estimated to be approximately Naira 70 billion, representingless than 0.5 percent of GDP. This is low comparedto other developingeconomies, although not unusual in Sub-SaharanAfrica3'. The share of mortgage loans and advances inthe banks' loan portfolios is less than 4% and these loans are concentrated in large urban areas like Lagos, Abuja andPortHarcourt. The Government's current framework for promotinghousingfinance hasbeenlargelyineffective. After a phasewhere the provisionofhousingfinance mainlyreliedon a state housingbank, inthe early 1990's the Governmenttriedto foster the emergence ofthe mortgagemarket by establishing 35The ratio is 15% in SouthAfrica and Chile 64 a specialized circuit composed of specialized lenders, Primary Mortgage Institutions (PMIs), a second tier entity, the Federal Mortgage Bank of Nigeria (FMBN) and a dedicated provident fund, the National Housing Fund(NHF). PMIs have failed inachievingtheir mandate which was to facilitate mortgage financing by distributinglow interest loans funded by the NHF. Total PMI assets represent less than 4 percent of total banking assets and mortgage advances account for only 23 percent o f consolidated PMI balance sheets. The majority o f P M I assets are held has deposits with other banks. Most stand-alone PMIs suffer from weak capitalization, little capacity to collect deposits, governance and skill deficiencies. Contributions to the National Housing Fund (NHF) resemble a regressive tax and their deployment has failed to promote long-term housing finance. The proceeds o f the 2.5 percent NHF contribution, given the low loan maximum amount and the absence of income related lending criteria, are effectively only used by higher income earners. As a result low income savers subsidize interest payments on mortgages for higher income groups. The impact of the NHF on development of housing finance has been minimal. NHF's cumulative disbursements over 13 years36have beenonly N5 billion. This i s a result both o f NHF's small fundingbase and long delays between application and fund allocation (up to 2-3 years). Re-alignment o f the activities of the National Housing Fund (NHF) to support public good provision is advised. NHF contributions have the characteristics of a specific tax levied for the needs of housing and could be usedto capitalize and finance the operation o f the various types of liquidity facilities and support o f lower income groups. Alternatively, the NHF funds could provide a basis for targetedsupport of socially-vulnerable income groups through specific income contributions for housing developments. Such income subsidies are superior to interest subsidies, as they can be targeted directly at low income households and do not distort financial market development. Vouchers to part-finance purchase of real estate are an effective way to administer such income subsidies. Banks' increased capacity and focus will provide new impetus to the finance o f housing. The entry of strongly capitalized banks into the market leaves little justification for a specific and privileged framework for PMIs. Going forward, unless they are able to identify an otherwise underserved market niche, PMIs are likely to operate as mortgage finance subsidiaries of commercialbanks. Continued access to cheap fundingwill most likely result inweak PMIs being kept afloat without adding value to the market. The prudential framework for PMIs needs to be strengthenedto create a level playing field for all actors inthe mortgage markets and to promote necessaryconsolidationand strengthening of the sector. Remittances are a major source of informal housing finance. Micro-level surveys suggest that a significant share o f these remittances are channeled into housing development - up to 50 percent of total remittance flows - thus dwarfing the current financing contribution provided by local financial markets. Thus, leveraging remittances as equity or security through a functioning mortgage market would provide substantial opportunities for kick-starting housing finance in recipient communities. Developing payment systems and products to provide incentives for a larger share of informally transmitted remittances to be available for intermediation through the financial system will improve the availability o f remittance finance for housing developments, see Section 3.d above. Innovative products that specifically target expatriates, like premium accounts with online access or combining mortgage products with real estate management services, could provide a platform to attract a larger proportion o f remittance flows. 36And N 8 billionfor developer finance 65 Long-term finance for housing in Nigeria i s constrained both on the origination (primary) side and the funding (secondary) side. Financial institutions are reluctant to extend long maturities to mortgagees because of a weak lending and operational framework and simultaneously lack the ability to access long-term financing to fundmortgage portfolios on their balance sheets. Primary Mortgage marketframework Banks are reluctantto develop mortgage financing capacity, as the mortgage lendingenvironment i s inadequate. Housing loans are large relative to the income of individual borrowers. Financial institutions rely on real estate as collateral so as to mitigate their credit exposure. Weak legal protection of secured property rights and enforcement of collateral, incomplete land titles and registrationas well as landtransfer restrictions hamper the development of housing finance. Land TitleRegistrationand CadastralMapping. The reform the LandUse Act o f 1978 is crucial for the development o f mortgage finance. Mortgage finance relies on the ability to transfer property ownership expeditiously and with limited costs. The provisions o f the Land Use Act - in particular the requirement that the Governor's consent to title transfers - result in significant delays and costs. In many states the fee income generated out o f the Governor's consent is a significant source of income. As a result, transaction costs for land transfers and registry of mortgages are prohibitively high. Consent fees amount to 15 percent or more of property valuations, and total stamp duties, registration fees and other levies can amount to 30 percent. These very hightransaction costs are a significant deterrentto the formalization of property rights -especiallybylowincomeholdersofcustomarytenurerights-andhamperthedevelopmentofa secondary property market. They provide a significant inducement for corrupt practices. Reform of the Governor's consent process in the FCT and Lagos state are steps inthe right direction and should be replicatedby other states, see Box 2. Box 1: Pilot reformsof the property registrationand cadastralmapping process Lagos state has halved aggregate cost from 30 percent to 15 percent o f the property valuation, and the Governor has delegated the power to administer the consent to several land commissioners and implemented a turn-around time o f 30 working days. The FCT has gone fiu-ther by establishing a streamlined and transparent system based on fully digitalized landregistries, and reducing consent fees to a low, nearly nominal level. This is a significant factor in encouraging financial institutions to provide mortgages in the FCT within the framework o f the FMBNbond issue. Reform o f the legal basis for title transfer would thus stimulate replication o f this mortgage finance "pilot". The administration o f the Governor consent should be based on a transparent and objective process, including standardized conditions and criteria, which, if fulfilled, constitute a right to receive Governor's consent. The introduction o f standard, transparent criteria will reduce legal uncertainties, facilitate the delegation o f power and result inmore efficient administration. Contingent property rights create legal uncertainty. Private land titles are often made conditional upon the fulfillment of land use obligations and can be revoked if these obligations are unfulfilled. Such land use obligations typically include the obligation to develop newly acquired land within a certain time frame - often 2 years - and other restrictions with respect to price and maintenance obligations. There i s also uncertainty as to the circumstances under which such obligations are met, especially in case of land and building developments. Lenders are rightly reluctant to finance developments with conditional land titles and, in turn, land owners lack the financing to carry out the developments they are obligated to. Despite the experience inthe FCT and Lagos in establishing efficient land registry and cadastral mapping systems large portions of land and properties, including large urban areas in Greater Lagos and federally owned land, remain. For this reason, housing developments are often 66 financed through equitable (unregistered) mortgages with limited validity and seniority. The establishment of a cadastral mapping system, as well as a computerized registration system, providesa key advantage to the FCT in developinga local housingfinance market and provides the basis for the establishment of an efficient property tax system. The cost of mapping and establishing the property registrationsystem was US$ 2 million for the city of Abuja, and the mappingofthe reminder ofthe FCT is expectedto cost US$1million. Secured lendingJFamework and enforcement. The legal framework needs to be strengthened to support robust and effective foreclosure procedures. Enforcement of mortgage rights in Nigeria i s inefficientand costly. It can take severalyears for a lender to repossesspledgedpropertyfrom a defaulting borrower, and there are many opportunities for willful defaulters to delay or stall foreclosure through court injunctions and other means. Creditor rights and insolvency systems are important in promotingfinancialsystem stability, fostering greater confidence in commercial contracts and facilitatingthe managementand resolution of defaultrisk. Effectivecreditorrights and preservation of asset values incase of default are crucial to broadeningaccess to credit. The Government is contemplatingreforms in this area, but seems to be considering to target the benefit of the reforms to the benefit of the PMIs. It is important that a better balance between mortgagecreditors' anddebtors' rightsbe establishedfor the whole spectrumof lenders. There is also a needto strengthen capacity ofthejudiciary system to handle commercialdisputes and enforce investorrights in a timely and efficient manner. Once ajudgment has been passed, evictions are difficult to enforce. As a result, lenderspreferto retain titles of the propertiesthey finance until loans are fully repaid or use arbitration mechanisms instead of judicial action. Various initiatives in this area deserve support - such as the FCT's efforts to streamline court foreclosure proceduresusinga specialized court and to specify alternativearbitrationprocedures in mortgage contracts, and the practice in Lagos State to promote judges with commercial expertise. Real Estate Development. Lack of primary infrastructureand urban planningmake costs of land development prohibitive. Housingdevelopmentsoftencarry the full costs of developingprimary and secondary water, power and road infrastructure. Lack o f comprehensive and reliable urban planning frameworks increase uncertainty over land investments and put existing land developments at risk. The absence of building standards and standardized, uniform valuation procedures limitthe abilityto define bankablehousingstock. Developmentof real estatetargeted at the rentalmarketwill continueto playan importantrole for housing provision, especially in urban areas. A large share of current real estate-related lending i s directed towards buy-to-letinvestors. However, the short maturity of such lending (usually only 3 to 4 years) results in high repayment charges and can therefore only support real estate targeted at the high-end and expatriate market segments. Creation of Real Estate Investment Trusts may improve access to finance for rental housing, especially as prudentialregulationof pension funds precludestheir direct investment inreal estate. Inaddition, it will be necessaryto establish a clear and balanced tenancy regime, including strengthened ownership rights, to support the development of a stronger build-to-rentmarket. Measures designedto strengthenthe legal and institutionalframework for property development will be important. A condominium law, includingrules to ensure the coverage of maintenance cost and a framework for the off-plan purchaseof new unitsto facilitate financing of unitsbeforetheir completion, will be importantto stimulatingthe construction of new homes and ensuring that such homes are affordable for a broadersegment ofthe population. 67 Promoting long-termfunding for mortgageportfolios The lack of long-termfunding has hitherto limited the ability of banks to fund long-termrisks. While the deposit base of Nigerian banks is growing and they may be willing to increase their exposureto the mortgage market, banks' appetite for term lendingwill be constrainedby the lack a long-termfunding (maturity risk). The availability of term-fundingis expectedto improve due to the recently instigated pension reforms. The investment rules promulgatedby the Pension Fund Commission favor mortgage-related securities by setting a 30 percent maximum limit for this asset class, provided the securities are rated as investment grade. The development of investment instruments meeting pension fund requirements will be key to increasing the availability of long-term funding. Such reforms further weaken the rationale for NHF in supporting the housingmarket with artificially low interest rates, especially given NHF's weak track recordin fulfilling its mandate, as describedabove. Developing secondary mortgage markets is challengingand simplicity is key. The sequenced introductionof morecomplex solutions will allowthe development of criticalmass inthe primary mortgage market and provide the time to carry out needed reform of the legal framework, including strengthening creditor rights and bankruptcy regimes as well as introducing market standardsfor mortgagecontracts and servicingof securitizedportfolios. In the early stages of mortgage market development off-balance sheet financing of mortgage portfolios through securitization is unnecessary. Keeping mortgage portfolios on the balance sheet of financial institutionsinthe early stages of mortgage market development is attractivefor banks. Banks can diversify their balance sheet structure with good riskheturn assets and mortgage finance provides a catalyst for building-up adequate retail risk management capacity. The sale of entire mortgage portfolios through mortgage-backed securities will only become necessary, once the share of mortgages in bank balance sheets becomes too large, in particular with respectto their capitalbase.Forthe time beingthis will notbe a constraint. Banks and other mortgage originators can access long-termfunding by issuingmortgage bonds. Mortgagebonds, which are securedby priority claims on the mortgages portfolios on the bank's balance sheet, provide efficient vehicles for tapping capital markets and are less costly than securitization. The IFC is using partial guarantees to support the issuance of mortgage related bonds by a number of Nigerianbanks. These bonds have been well receivedby the market and there is scope for replicationo f such partialguarantee structures, providedguarantees are issued strictly according to sound eligibility standards relatingto the quality o f the issuing banks, the underlying mortgage portfolios etc. The design of the Federal Mortgage Bank of Nigeria's (FMBN) planned bond issue warrants further consideration, particularlyas regards the efficient use of guarantees, see Box 3 below. 68 Box 2: Design of the plannedFMBNbond issue The structure ofthe plannedFMBNbond issue provides little momentum for developing sustainable long- term funding for the Nigerian mortgagemarket. The FMBNbond issue addresses a very specific need-to promote the sale of government-owned houses in the FCT. A more systemic approach is needed to promoting the availability of long-term funding focusing on the policy framework for primary and secondary mortgage markets. The specific features of the FMBN bond issue, including the federal government guarantee and its eligibility as a zero risk weight, liquid asset for the purposes of calculating banks' liquidity and capital ratios makes it interesting for banks as investors, but do not stimulate access to long-term funding from non-bank investors. The incometax exemption applicable to the FMBN issue i s designed to lower the interest rate payable on the bonds, but may well be counterproductive in a development perspective, because it will reduce the appetite of tax-exempt pension funds to invest in those bonds. Such subsidies are broad-based and are therefore unlikely to be effective intargeting lower income families. Inaddition mortgage finance facilitation ofthis kind does not warrant provision of a full Government guarantee, given the underlying quality of the collateral. Market participants are already aware that similar guarantees are unlikely to be available for future bond issues. The Government might consider leveraging the support which it has committed for the bond issue by transforming the full guarantee into a partial guarantee -thus leveragingthe guarantee to facilitate a larger issue volume. The Government might also consider bringing in private sector institutions or IFIs as joint guarantors. The issuance of such Government guarantees should in any event be limited to the initial or start-up phase of mortgagebond market. Providedinstitutionalreforms are undertaken, the FMBNcould fulfill several catalytic functions to support long-term mortgage funding. The mandated reforms of the primary mortgage environment, the consolidation of the banking sector and the recently implemented pension reform should considerably augment the availability of mortgage funding. Given these reforms, pressure on FMBN to provide funding should subside, and FMBN will be able to focus on its core mandate as the apex institution for the housing finance - providing public support and reducingdistortions that hamper overall market development. Several examples of international best practice couldbe considered: o Provisionof liquidity for mortgage issuers that allows financial institutionsto access liquidity for limited time periods, for example, through repos or stand-by facilities, which are collateralized by mortgage portfolios that meet certain quality criteria. Such a mechanism would provide banks with insurance against that possible future liquidity risks andprovideincentivesto enhancematuritytransformation3'. o Provision of market liquidity for investors in mortgage bonds. The same facility couldbe usedto allow investors in pre-qualifiedmortgagebonds to rep0the bondsat a discount and thereby allow them to access liquidity or liquidate positions in the absenceof a liquid secondarytradingmarket38. 37 This facility would not re-financethe entire portfolio untilmaturity, as i s the case ina securitization, and thereforewould not require detailedvaluation ofthe underlying mortgage assets, which is an obstacle ina nascent marketwith a weak basis on which to assess future cash flows. The establishmentof eligibility criteria for mortgage portfolios will promotestandardizationas a first step towards developingcapacity for securitizationat a later stage. 38 Similar schemes have beenestablishedinsome nascentor emergingmarketslike Colombia andMexico. A possiblevariation would be to provide liquidity only after a certain time period, for example three or five years after the issue date, so as to make the facility availableonly to investorswith a long-term investment horizon. 69 o Support of lower income housing. The FMBNcould (i)provide guaranteesto grass root lenders or micro finance institutions to enable them to raise funds from the formal financial systems; (ii)establish a secondtier liquidity facility, which uses the proceeds from bond issues (possibly with partial support and risk-sharing by the Government or donors) to re-finance grass root lenders or micro finance institutions that do not have good access to longterm resources; and (iii) providedirect mortgage insurance on loans for low-incomeborrowers. All three options requirevery clearly defined eligibility criteria, transparent processes and adequate institutional governance arrangementsto avoidundue distortions and ensure that support reaches the intendedlow-incomegroups. Innovativeinstruments, suchas liquidity facilities and inflation-indexedsecurities, can instilltrust and increase provisionof term funding. The use of liquidity facilities with adequate financial backingwould supportthe Government's effortsto signal commitment to providinga sustainable macroeconomic framework. In addition, the promotion of inflation-indexedinstruments could offer attractiveprotectionfor pensionfunds and other long-terminvestors. For investors, such as pension funds, providing indexed instruments is often perceived as an important incentive for engaging in mortgage finance. For borrowers, inflation-indexedinstruments can pose issues, but are appropriate inas muchas real estate security provides a naturalinflationhedge. However, the successful introductionof inflation-indexedinstruments does depend on development of reliable measures of inflationand transparency as regards the objectives and implementationof monetary policy. 70 TECHNICAL ANNEX 5-C: DOMESTIC TERM LOANTRANSACTIONS IN NIGERIA Box 1. Examplesof domestic term loan transactions in Nigeria Oando Power / Gaslink In 2002, Oando Power raisedNaira 1.9billion ($14million) in 5-year project finance for its Gaslink subsidiary from 5 local banks. Gaslink had a 3-year operatinghistory and Oando Holding Company provided a corporate guarantee. Gaslinkis looking to raise additional5-year financingfor Naira 2.8bn ($20 million). MTNNigeria In 2003, after two years of operations, MTNNigeria syndicateda $395 million medium term facility to international and 15 domestic banks. The $250 million domestic portion was designed to repay 25% in year 3, 25% in year 5 and 50% inyear 7. At the end of each period, lenderscould elect to roll over or redeemtheir debt. IFC provideda standby facility for redemption. MTNrecentlyraisedan additional$120million equivalentfrom domestic banks. Obajana Cement Plant (Dangote Industries) In 2005, this project required debt financing of about $375 million. $150 million equivalent was provided by a syndicate of 15 local banks with a maturity of 3 years. The remaining$225 million were raised through term loans from IFC andEIB with guarantees from exportcredit agenciesandinternationalbanks. VMobile Nigeria In December2005, Virgin Mobile Nigeria raised $1.1 billion in a combination of domestic and external finance. 16 domestic banks provided a total of $450 million equivalent with a term of 6 years, covering the funding horizon of VMobile's 5-year investment plan. The $700 million in external finance are being provided in 2 phases, include vendor finance andare partially supportedby Export Credit Agencies. Nigeria Flour Mills Of total project costs of $420 million, $205 million equivalent are currently beingraised locally through a syndicate of 8 Nigerianbanks. Two Naira-denominatedtranchesare offeredwith a 7 and 9 year maturity and a2 year grace period. This transactionprovidesthe first 9-year loantranche inlocal currency. LekkiTollRoad, Lagos This pilottoll roadinadensely populatedarea inLagos is projectedto be financially viable and is lookingto raise $145 million in 15-yearsenior debt financing combinedwith $105 million inequity andmezzaninefinance. Local banks are structuringa 10-year term loan with a cash sweep which could reducethe tenor. Banks are initially expected to lend for 5 years anda standby facility is beingcontemplatedfor banks looking for repaymentafter the initial period. 71 TECHNICAL ANNEX 5-D: TECHNICAL NOTE ONA PROPOSED INFRASTRUCTUREFACILITY FORNIGERIA Context Acting on its strategic priorities for growth defined in NEEDS, the Nigerian Government has initiated far-reaching reforms in the financial and infrastructure sectors. In the financial sector, 25 large banks have emerged from successful consolidationwith a strong capital and deposit base which can be leveraged for increased credit to the private sector. Similarly, pension and ongoing insurance reforms have created growing long-term savings as a natural source for long-term investment. The existing potential of Nigeria's financial system for term finance is estimated in excess of $10 billion (see Table 5-2 in Chapter 5 o f the main report). In infrastructure, government has undertaken a number of initiatives to improve the availability, quality and governance o f infrastructure services, including measures to attract private participation in the electricity and transport sectors. To secure the benefitand irreversibility of reforms, further progress is required infinancial sector and infrastructure development. Inthe financial sector, banks need to build capacity in assessing and managing long-term credit risk, while capital market development i s crucial to provide refinancing for bank loans and investment opportunities for pension funds and other institutional investors. In the infrastructure sectors, increased private sector participation will depend on transparent and predictable sector frameworks and improved access to long-term finance. This requires a clear definition of public versus private sector roles in infrastructure through regulation and inthe context o f a systematic investment planning and project prioritization process. Local currency financing for infrastructure provides opportunities to combine financial sector development with improved private participation in infrastructure, with ensuing benefits for economic growth. Benefits of local currency financing for infrastructure include Improving the absorptive capacity of the economy: Investments in supply side productivity such as infrastructure can alleviate inflationary pressures created by excess liquidity from e.g. export revenues. Infrastructure projects provide opportunities to channel domestic savings to productive uses. Creating investment opportunities for domestic banks and pension funds: For their increased capital and funds, banks and pension funds are looking for long-term assets with appropriate returns. If appropriately structured, public-private partnerships for existing infrastructure assets can provide creditworthy lending and investment opportunities with relatively predictable long-term cash flows. 0 Improving governance: Private participation in infrastructure reduces the scope for government intervention, thereby increasing efficiency. Domestic finance for private project sponsors signals government commitment to reform, as government failure to honor its obligations bears the cost of destabilizing local financial markets. Eliminatingforeign exchange risk Matching local currency revenue streams with local currency liabilities reduces vulnerability from exchange rate movements. Experience in the 1990's has shown that material foreign exchange fluctuations caused many contract re-negotiations for local currency earning infrastructure projects which were financed with dollar based internationalloans. 72 0 Reducingfinancing costs: Domestic lenders and investors are better equipped to assess and price political and other government-relatedrisks. Consequently, they are willing to provide longer maturitiesandrequirelowerriskpremiaand less risk mitigationcompared to internationallenders. This can reduce financing costs, thereby enhancing financial sustainabilityof infrastructureprojects and loweringuser fees. Access to domestic term finance for infrastructureis currently limited by several structural and riskfactors in financial and infrastructure sectors: Ininfrastructuresectors, risksfroma privatelenders' perspective include (i) High regulatory risk as sector regulators, tariff regimes and market rules are still beingdeveloped; (ii) Highpolitical risk given a historyof governmentnon-paymentto utilities, ongoing government interference in infrastructure sectors, the lack of transparent public investment planning including a defined framework for government support to PPPs, and unclear institutionalresponsibilities; (iii)Highcommercial/demandrisk dueto e.g. significantcommercialandtechnical losses in the electricity sector and an untested user willingness to pay tolls in the roadsector. Inthe financialsector, risksandimpedimentsto term finance for infrastructureinclude (iv) Relatively high interest rate risk as financial markets currently do not offer fixed rateproducts and interest rate volatility has historicallybeenhigh; (v) Highrefinancing and liquidity risk as capitalmarketscurrently lack corporate bond productsfor bank loanrefinancingand liquidity for take-outfinancing; (vi) Barriers to entry to capital markets due to high costs of issuance and onerous reportingrequirements; (vii) Restrictionson pension fund investments in infrastructureprojects resultingfrom minimumratingand listingrequirements. An infrastructurefacility for Nigeria is recommendedto mitigatethese risk factors inthe context of continuing financial sector reforms. Over the short- to medium-term, such a facility could address the current risk factors and market failures througha combinationof risk mitigationand financing instruments. The medium term development goal of the facility would be to enhance access to market-based financing by providing incentives for and supporting financial sector reforminrelatedareas. Objectives The objectives ofthe proposed infrastructurefacility for Nigeriainclude Creatingfiscal space through improved access to private investment in infiastructure: To limit fiscal costs to Government, the facility would aim to provide a standardized framework for private participation in infrastructure. Elements of such a framework include(i) rigorous economic and financial cost-benefitanalysis for projectsselected for private participation, (ii)consistency with priorities identified in the Government's Medium-TermExpenditure Framework and sector strategies, (iii)limiting government 73 support39to projects which are not financially viable on their own, and (iv) developing a transparent approach to government support for PPPs. In the case of guarantees, government exposure should be limitedto risks which are under government control. Supporting the irreversibility of inJFastructure reforms through domestic financing of benchmark transactions: Government has embarked on a comprehensive program for private participation in infrastructure. If all o f the projects in the pipeline materialize within the anticipated timeframe, significant private financing is required. Access to domestic finance will be triggered by successful benchmark transactions. Such transactions initially require risk mitigation which can be scaled back over time as financial sector reforms progress and the availability of market-based financing increases. Promoting financial sector development: The facility would aim to promote local currency financing for infrastructure projects in the context o f broader financial sector reforms focused on i.Establishingapricingbenchmarkforlongertermdebt ii.Extendingtenorsofbankloanstoproductivesectors iii.Facilitating long-term investment by local institutional investors in debt and equity instrumentsof infrastructure projects iv. Developing primary capital market issuanceof corporate bonds, v. Developing and improving secondary capital market liquidity, vi. Increasingthe penetration of credit ratings. Providing associatedtechnical assistance inrelevant areas, such as i. Principlesandmethodologiesofinfrastructure investment planningandproject selection criteria for Public-Private Partnerships, ii.ReviewofthelegalandinstitutionalenvironmentforPPPsinNigeria, iii.Guidance for infrastructure project preparation - working with all relevant government agencies, complementing activities o f the proposed Nigeria InfrastructureAdvisory Facility developed by DFIDas required, iv. Review of barriers to capital market development, such as issuance costs and investment restrictions, v. Training for commercial banks and pension funds in project finance, financial structuring and long-term credit risk analysis and management, vi. Capacity building for regulators in long-term credit risk analysis, management and supervision. As outcome indicators, the facility would aim to (i)increase the share of private and local currency financing of infrastructure projects in Nigeria, (ii)bring a target number o f projects to completion within a defined time frame, and (iii)facilitate a minimumnumber of project bond issues indomestic capital markets. ''Governmentinvestmenthkindcontributions,direct subsidiesor guarantees 74 Key principles Basedon lessonslearnedfrom experiencewith infrastructurefunds and infrastructurefacilitiesin other countries, a few key principlesneedto be taken into account when designingthe proposed infrastructurefacility for Nigeria. Infrastructure funds versus infrastructure facilities: Infrastructure funds are established for the commercial financing of a limited number of identified infrastructure assets. Privateandpublicfind investorscommittheir moneyupfront andprivatesector investors have minimumreturnexpectations over a definedtime horizon. Funds typically have a limitednumber of financing instruments, investingeither indebt or equity, and a broader definition of eligible infrastructure sectors. Infrastructure finds are suitable if (i)an identified pipeline of projects for immediate investment exists, (ii)projects have sufficient cash flows to meet return expectations and (iii)investors are experienced in project finance. To accommodate a degree of uncertaintyaround the timing and return structure of the planned PPP projects in Nigeria, a more flexible facility approach is recommended. An infrastructurefacility does notrequirepre-committedprivate funding, as it can provide risk mitigation and possibly donor financing as projects become available. This avoids the pitfallsof minimumreturnrequirementswithin a defined time horizonas well as portfolioconcentrationrisk. The importance of a supportivepolicy environment: Infrastructurefacilities can provide short-term risk mitigation and complementary funding to trigger access to domestic private finance, while they cannot substitute for progress in financial and infrastructure sector reform. Infrastructurefacilities in other countries have not met their objectives where government commitment to comprehensive reform was lacking. In addition, the facility should be viewed as transitional to avoidpermanent subsidies. An infrastructure facility for Nigerianeedsto be embedded in (i) broader reformprogram for financial an sector development, (ii)ongoing progress in implementing infrastructuresector reforms including the completion of regulatory frameworks, (iii)a transparent and fiscally prudent approach to government support in infrastructure PPPs, and (iv) a sound legal and institutionalenvironment for PPPs. The crucial role of strong, private sector-led governance structures: Infrastructure facilities require an independent, competent and impartial governance structure. Experience with World Bank supported infrastructure facilities has demonstrated that public sector entities generally lack technical capacity and independence from political influence to adequately select, prioritize, prepare and market projects for private investment. Existinginfrastructurefunds, on the other hand, are exclusivelymanagedby private sector fund managers. In line with best practice standards, private sector management is crucial to gain access to necessary technical knowledge, build capacity andreducethe potentialfor governmentintervention. The need for a thorough demand assessment: Experience with the implementationof World Bank facilitated infrastructure projects shows that facilities have at times been supply driven, while not enoughattention was givento assessingthe potentialandtiming of demand. A thoroughdemand assessment is therefore requiredto align risk mitigation and financing instruments with the size and timing of future transactions. To date, a preliminary demand assessment for Nigeria indicates that at least 6 PPPs will likely be looking to raise financing in the next 12-18 months4', while up to 40 PPPs have been 402 distributionconcessionsand 4 railway concessions;oil company IPPs are excluded as they will have a separate financingstructure andthe Abuja airport concessionwill not requirecredit enhancement. 75 mentioned by various government agencies for potential realization over the medium term. The World Bank is financing transaction advisors in the electricity and railway sectors who will assist in sector restructuring, conduct a detaileddemand assessmentand assist in structuring PPPs in the next 12 months. Compared to other facilities the potential demand for an infrastructure facility is relatively higher in Nigeria combined with larger potential of the domestic financial sector to provide term financing for infrastructureiffacilitatedby appropriatecredit enhancement. 0 The importance of a systematic approach to risk allocation: Successful infrastructure funds such as the Korea Road Fund have a clearly defined risk sharing between the public and private sector and a standardized policy for government support to PPPs which declinedover time. This increasedaccess to privateinvestment andreducedreturn requirements. A systematic, transparent and predictable approachtowards risk allocation between the public and private sector is a pre-conditionfor maximizingthe number of private bidders and effecting a reasonable transfer of economic and financial risks at limitedfiscal cost. Recommendationsfor Nigeria a) Instruments To achievethe development objectiveof improvingaccess to privateand localcurrency financing for infrastructureprojects, the proposed infrastructurefacility would provide a combination of riskmitigationand financinginstruments,possiblyin a phasedapproach. Risk mitigationinstruments can enhance access to local currency financing for projectsready to be implemented. Recommendedrisk mitigationinstrumentsfor Nigeriainclude: 0 Partial Risk Guarantee Facility: Partial Risk Guarantees (PRGs) cover debt service defaults on loans to a private sector projectresultingfrom Government's failure to honor its contractual obligations. PRGs can backstopvarious forms of government obligations, such as implementation of untested regulatory regimes, minimum revenue guarantees, government off-take commitments and other government payment obligations. In Nigeria, PRGs could, for example, provide risk mitigation (i)to commercial banks against default resulting from breach of contract by government in projects with significant government involvement, such as toll roads concessions, newly regulated electricity generationprojects andelectricitydistributioncompanies, and(ii) to privatized companies against revenue shortfalls resulting from government failing to adhere to agreed regulatory structures, such as tariff increases. Guarantee amounts typically only cover a defined portion of total project debt to attract private financing while limiting contingent liabilities for Government. Benefits of PRGs include (i)enhanced credit ratings, improving eligibility of infrastructureproject bonds for investment by pension funds, (ii)extended maturities, improving access to longer-term financing and (iii) reduced financing costs. The PRG facility would be offered by the World Bank and would require a counter-guaranteeby Government. 0 Liquidity facility: Nigerian's capital markets do not yet provide long-term capital financing as a take-out for medium-termbank loans and liquidity for exiting securities investments. A standby liquidity facility would help extend loan maturities from local banks by mitigating refinancingrisk. Such a liquidity facility would provide lenders a refinancing and/or redemption option after an agreed term and in the absence of refinancingpossibilities in local capital markets. The size and support of the facility 76 could decrease over time, as depth and liquidity in capital markets improve, especially following implementationof pensionand insurancereforms. Government could consider dedicating a portion of its oil funds for this purpose.4* Government backstopping or financing of the liquidity facility would signal commitment to sound monetary policy, enhancingcapital marketconfidenceand development. Ifneeded, the infrastructure facility could complementriskmitigation instruments with financing instrumentsfor which market-basedfinancing is not yet available, such as subordinateddebt and equity finance. 0 Subordinated or mezzanine debtJinance: Nigeria currently has limited local sources of equity to invest in infrastructure projects. Depending on the number of infrastructure projects in search of financing at any given time, this could result in equity funding shortfalls. To leverage existing equity investors, a donor-initiated facility could provide subordinated debt or quasi-equity instruments. The facility could provide debt instrumentsina project's capital structure which the private sector inNigeria does not yet offer, such as mezzanine finance. Over time, such a donor-initiated facility could grow into a full-fledged commercial mezzanine fund, as a growing number of infrastructure projects provide investment opportunities and domestic investors become more familiar with this asset class. 0 Equity Jinance: To date, private equity in Nigeria's financial market is limited and provided for a term below the financing requirements of infrastructure projects. If a sufficient number of infrastructureprojects materialize at the same time, a passive equity fund might be considered to complement domestic sponsor equity. The fund could initially be donor-facilitated and over time be phased into a full-fledged private equity fund for infrastructure inNigeria. b) Institutionalarrangements To avoid the risks associated with public sector management and reflecting international best practice, it is recommendedthat the infrastructurefacility be managed by an independent, private facility manager which would be selected on a competitive basis. For project preparation, the facility manager would work in close cooperation with all relevant government agencies, including Ministry of Finance, infrastructure line ministries, BPE, Infrastructure Concession Regulatory Commission and others. In addition, the facility manager would coordinate with the proposed DFID Nigeria Infrastructure Advisory Facility to ensure an integrated approach to projectpreparationand financial structuring. IDA and potentially IFC and MIGA involvement in the facility bring the additional benefit of World Bank Group governance standards. The IDA Partial Risk Guarantee Facility could be provided as an adaptable programmatic loan to be reviewed every two years to scale the facility to actual demand. IDA approval would be sought for a list of pre-identified projects to be supported by guarantees, with individual project approval for larger scale projects beyond a defined transaction amount. To ensure immediate facility utilization, facility approval would be sought from IDA jointly with approval for the first concession projects that will benefit from support underthe facility. ~ ~~ 4'The political economy considerationsofthe required approvalby the National Assembly ofthe use ofthe oil funds would needto be carefully considered. 77 The facility would have pre-defined and standardized eligibility, evaluation and performance criteria. The facility would aim to support those projects which are economically viable but not financially sustainable without access to risk mitigation or additional financing. c) Technicalassistance The facility would include dedicated technical assistance to financial and infrastructure sector stakeholders in areas which are crucial for successful implementation. Potential areas for technical assistance(and recipients) include: i.Principles of infrastructure investmentplanning and project selection criteriafor Public-Private Partnerships (Ministry o f Finance, infrastructure line ministries,BPE) ii.ReviewofthelegalandinstitutionalenvironmentforPPPsinNigeria, includingthe institutional structure to support the proposed facility. iii.Infrastructure project preparation - guidance for relevant government agencies (infrastructure line ministries, BPE, Infrastructure Concession Regulatory Commission) iv. Review of barriers to capital market development, such as issuance costs and investment restrictions(Pension Commission, Nigeria Stock Exchange), v. Training in project finance, financial structuring and long-term credit risk management (commercial banks, pension fund administrators) vi. Training in long-term risk management and supervision (Pension Commission, Insurance Commission, Central Bank as necessary). Figure 1 Nigeria Infrastructure Facility Public Sector Infrastructore Facility Private Sector I I I pendent fund mana I I 78 Table 1: Infrastructure Facility for Nigeria - Indicative Key Features and Terms4* Overall Facility Structure -applicable to all tranches Facility Structure Public-private facility comprised o f one or several o f 1. Partial RiskGuarantee Facility 2. LiquidityFacility 3. Subordinated debt facilityhnd 4. Private equity fund Scope Local infrastructure assets in identified infrastructure sectors [electricity generation, distribution and transmission; road; railway; aviation; port; telecom; water] under a Public-Private Partnership arrangement [concession or other contractual arrangement for private financing, construction, rehabilitation, operation and maintenance o f infrastructure asset] Project eligibility criteria 1. Infrastructure project under a PPP contract awarded in a transparent and competitive manner 2. Economically and technically viable, but not financially viable without facility support 3. Compliance with government policies as identified in NEEDSKPS and relevant infrastructure sector policies 4. Minimumequity finance of20% oftotalproject cost Facility manager Competitive bid for independent,private facility manager Facility size TBDbasedon detailed demandassessment Availability period The facility will have [3] successive availability periods o f 2 years during which projects can be submitted for consideration and support Tenor o f Underlying The maturity o f the underlying financing guaranteed or supported by the Infrastructure Project Facility cannot be less than [7] years for debt financing and [5] years for equity Financing financing 1. PartialRiskGuarantee Facility Nature o fthe Guarantees Partial Risk Guarantees to commercial lenders o f an infrastructure project for the event o f default resulting from government non-compliance with specific contractual obligations Facility size TBD based on detailed demand assessment, reviewed every 2 years Individualguarantee Guaranteed amounts are limitedto the lower o f [50%] o f infrastructure project coverage debt or [US$50] millionequivalent Currency o f guarantees Guaranteescan be issued inNaira or US$. Ifissued inNaira, a maximumUS$ equivalent amount can apply. Nonre-instatability Once a guarantee has been triggered and IDA has effected payment, its effect stops i.e. the guarantee can not be reinstated. Non-accelerability Payments by IDA to a particular infrastructure project will be limited to scheduled debt service payments as they become due I D A Approval Projects above target size [tbd]: individual IDA guarantee approval Projects below target size [tbd]: IDA facility approval 2. Liquidity Facility ~~~ Nature o f facility I a) Refinancing facility for local commercial bank loans to infrastructure projects b) Take-out financing for debt capital markets investors at prevailing market 42 Information in [brackets] is subject to further discussionand will be finalizedjointly with the government 79 + I Ivera11Facility Structure-applicableto all tranches prices c) Take-out financing for local private equity investors at prevailing market prices Eligible beneficiaries Domestic commercial banks, institutional investors (pension funds, insurance I companies) andpassive equity investors in local infrastructure projects Facility size ISource Government / IF1 3. MezzanineFacilitymund Providing subordinated debt instrumentsto infrastructure projects Local infrastructure projects which require subordinated debt or quasi-equity finance to attract senior debt finance from local commercial banks andor senior bond finance from capital markets/institutional investors Facility size Initially donor facilitated, can grow into a stand-alone Mezzanine Fund over time 4. Private Equity Fund Nature o fthe fund Passive equity investments Eligible projects All local infrastructure projects which can generate adequateequity returns Facility size [TBDI Source International and local institutional investors, including banks, pension funds, insurance companies, private equity funds 5. TechnicalAssistance Potential areas for TA 1. Principles of infrastructure investment planning and project selection 2. Analytical tools and decision criteria for PPP prioritization and preparation, 3. Review o f the legal and institutional environment for PPPs inNigeria, 4. Review o f barriers to capital market development, such as issuancecosts and investmentrestrictions, 5. Institutional capacity building for all relevant stakeholders, such as commercial banks, pension fund administrators and financial sector regulators. 80 Table 2: Indicative projectpipelinefor infrastructure PPPs in Nigeria Sector Project Indicative timing Electricity Geometric Power 2006 4 IPPs(Shell, Mobil, Chevron, Agip) 2007 2 PHCNdistribution companies(Abuja, Ikeja) 2007 4 remaining PHCN distribution companies 2008 11PHCNgenerationcompanies 2007/2008 7 power stations inNiger Delta 200812009 Aviation Abuja airport 2006 4 other international airports 2008 Railways Concessionof Central Line 2007 Concessionof WesternLine 2007 Concessiono fEasternLine 2007 Concessiono fNorthernLine 2007 Lagos UrbanTransport Concession 2008 Rail connections to ports TBD Road Lekki toll road, Lagos 2006 Pilot highway toll road 2007 Planned concessions of 15 new roads and bridges under TBD BOT scheme Planned concessions of 15 existing roads and bridges TBD underRMOT scheme 81 8 8 . 8 8 8 I I d B d B .-0E 3 'Z 0m FA U 8 8 m h 2 F P 3 8 m 10 i Ce 3 @ U 3 MJ 9 U 7 8 3 v) 0 0 c\1 .9 I Ia- r I t- .e 3+G I I InfrastructureFundsversus Infrastructure Facilities There are some fundamental distinctions between an infrastructure fund compared to an infrastructurefacility. Funds are set up for commercial purposes where investors commit their money upfront with minimum return expectations over a limited time horizon. In contrast, facilities serve more developmental purposes as credit enhancement andor financing is made available as andwhen projects materialize. An infrastructurefundhasthe followingfeatures: Return requirements: While infrastructurefunds can support developmentalobjectives, they needto fulfill commercial objectives. Privatefund investors committheir money up front and havereturnexpectations over a defined time horizon. As a result, they evaluate the timeliness and speed of the investment pipeline materializing and its projected cash flow streams. Return expectations are typically 15 - 20% for equity investments in infrastructure-relatedactivities. Investment objectives: The certainty of the project pipeline is critical to the success of raisingcapitalfor a fund. The fundaims to invest inor acquire a portfolioof quality and diversifiedinfrastructureassets over a relativeshort periodoftime (3-4 years) which will provide investors with a combination of stable cash distributions and capital gains. Access to a wide range of investment opportunitiesis required for investors to diversify their portfolio. 0 Financing instruments: Funds are typically structured with a limited number of instruments and generally in one security category. Debt funds and equity funds attract different sets of investors with different risk profiles as well as different return and investment requirements. Similarly, fund managers have specific skills and expertise. Equity funds are more prevalent as returns are higher, providing more upside potential and supportingthe costs ofthe funds moreeasily. 0 Pre-determination of target investors: Typical equity fund investors are local institutional(pension funds, life insurancecompanies) and financial investors, as well as industry participants. Historically, infrastructure fund investors have been offshore investors.The limited number of infrastructuredebt funds have a significant amount of support from international financing institutions, donor agencies or governments. For example, the EmergingAfrica InfrastructureFund's equity investors are donor agencies which account for 1/3 offund capital. Institutional structure: All funds are run by an independent private fund manager who managesthe fundthrougha special purposeasset management company. Fundmanagers are key to fund performance as they identify, screen and evaluate investment opportunities, recommend investments to the fund's board and support the board in its decision making. It is common practice for the fund manager to have significant experience inthe fund's targetedsectors. Fundmanagersare generallyselectedthrough a competitiveprocess,engagedthrougha managementcontract with the fundandrewarded basedon fundperformance. 86 Governance: Independent governance of the fund is an important factor for investors. The Board should represent interests of the investors in the fund and monitor the performance of the fund manager. The fund's governance should be overseen by the Boardwith the advice andrecommendationofthe fundmanager. Limited size and life: The funds all have a minimum size and life by which all assets must be divested and the fund terminated. The average size range of infrastructurefunds is $200-$500 million. This is because historically, small sized funds performed worse than larger ones. Larger funds are able to attract more experienced managers, invest in sizeable assets or acquire significant equity stakes, cover management costs or spread risks adequately and created a more expanded diversified portfolio. Equity funds typically have an 8 to 10 year "life". This means that the funds should be fully invested inthe first 3-4 years andthe fundmanagermust lookto start divestingits equityholdings by the 6* year. Role of the international donor/development community: Many infrastructure funds have been leveragedwith funds from internationalfinancial institutions. For example, OPIC, as a US. government agency provides an AAA loan guarantee on debt borrowed by approved equity funds. Its guarantee can support debt for up to one third of the fund's total size. Incontrast, infrastructurefacilities providea moreflexibleapproachandlendthemselves betterto achievingthe development objectives for Nigeriadueto the followingfeatures Objectives and return requirements: Facilitiesare accessible on a needsbasis as projects materialize. Therefore they can accommodate a higher degree of uncertaintyregarding the timing of realizingspecific transactions. Providers of facilities are often IFIs/donors which supportdevelopmentalobjectives over returnobjectives. Financing instruments: Facilitiescan combine a number of financingand risk mitigation instruments, includingsenior and subordinated debt finance and equity finance as well as liquidity support and guarantees. The majority of IFI-facilitated facilities have historically focused on supporting additional debt finance, through e.g. mezzanine finance or debt guarantees. Some facilitieshave includedequity finance. Absence ofportfolio concentration risk: As facilities supportthe realization of individual projects, investors can chose their sector exposure. This potentiallyallows for better risk diversificationcomparedto a fund approach. 87 StatisticalAnnex Tables 88 g l i 0 :y N -$ 0 0 N 8 0 N N 0 O N m P - 0 e4 3 0 0 N 0 0 W m m e " 0 9 1 1 0 N QI QI 9 0 QI 9 t- QI 53 W QI 9 vr QI 9 "i 2 2m m Z?Z d m - ? z s Z N m ? f r - m m - NN q = O ! s o 2 m 0 - 7 9 G o m 0 M o m w ? Q \ 0 9 - o ! F + e - 2 2 2 2 N N 0 4: d d 3 - 4 0 v ; w o 2 0 "O2 l - w W r2 o\ m -r-: l- W `c! W 3 o\ 2 W W 9 r: 3 3 00 m W m W N \l 4 ? 0 r: - 0 0 0 W 2 W l - md W l- W d 2 2 s 03 m 0 W m 3 z 9 s w o-\ 0 2 3 N W 0 N W 0 m W 3 `c! W z l- l- o\ N 4: w -4 0 l- W W f N 3 2 2 o\ i N d 2 -4 3 x N 2 m 00 a9 -05 40 s a h 0 V .C UB a IE ,, I I >>I \ \ \ < a \ .( h 3 2 * h n ? 9 9 9 2 n - W W a m m b Table 5: Nigeria Social Indicators Latest single year Same regionlincorne group Sub- Saharan Low- 1980-85 1990-95 1998-2004 Africa income POPULATION Total population, mid-year (millions) 78.4 103.9 128.7 725.8 2,343.0 Growth rate(%annualaverage for period) 2.7 2.8 2.3 2.3 1.9 Urbanpopulation (YOof population) 30.7 39.5 47.5 36.4 30.6 Total fertility rate (birthsper woman) 6.9 6.4 5.6 5.3 3.7 POVERTY (96 ofpopulation) Nationalheadcountindex 43.0 34.1 Urbanheadcountindex 31.7 30.4 Ruralheadcountindex 49.5 36.4 INCOME GNI per capita (US$) 390 230 430 601 507 Consumer price index (2000=100) 3 57 176 118 120 Foodprice index (2000=100) 3 64 153 INCOME/CONSUMPTION DISTRIBUTION Gini index 43.7 Lowest quintile (%of incomeor consumption) 5.0 Highest quintile (%of incomeor consumption) 49.2 SOCIAL INDICATORS Publicexpenditure Health(% of GDP) 1.3 2.4 1.3 Education(YOof GDP) 0.9 3.4 3.1 Net primaryschool enrollment rate (% of age group) Total 88 64 79 Male 95 68 82 Female 81 60 75 Access to an improvedwater source (Y6 ofpopulation) Total 49 60 58 75 Urban 78 72 82 89 Rural 33 49 45 69 Immunization rate (% of children ages 12-23 months) Measles 17 44 35 64 64 DPT 16 34 25 64 67 Child malnutrition (%under 5 years) 39 29 43 Life expectancy at birth hears) Total 46 45 44 46 59 Male 45 44 43 46 58 Female 48 46 44 47 60 Mortality Infant (per 1,000 live births) 120 120 101 100 79 Under 5 (per 1,000) 232 230 197 168 122 Adult (15-59) Male (per 1,000 population) 535 476 504 489 300 Female(per 1,000 population) 453 401 494 467 246 Maternal (modeled, per 100,000 live births) 800 921 682 Births attended by skilled healthstaff (YO) 31 35 42 40 Note: 0 or 0.0 meanszero or less than halfthe unit shown.Netenrollment rate: break in series between 1997 and 1998 due to change from ISCED76to ISCED97.Immunization:refers to children ages 12-23 monthswho receivedvaccinations before one year of age or at any time before the survey. World DevelopmentIndicatorsdatabase, World Bank - 5 April2006. 93 . . . . . . . . . . .. .. .. .. .. .. .. .. . . . . . . . . 3 s w m m b \ o r - o w w m o w o m o m b w m b m - w m o m o o o m m - - N N N N b N w r - m m m w c - h ( h ( m ze ze V U FI 0 e0 a 3 cL cL 0 0 0 0 9 9 9 9 q q o q : - O W b W w N N N -mmm N m N 0 0 - 0 0 o o - N 2 w - - 0 m m o m r m o w u O N O P m m l - 4 7 m w I n - m w C h 3 m o m u w o w c w e 4 0 7 W N N T L0 m o m a o m o r I n o w u Q \ m C l c : 2m 2 W x I- 2 3 2 00 - p! m 2 m t': d m 2 W m 3 I- Y 6) P, IA W m I- O 0 m - 0 m d o o m d N N 1 4 0 . N N O O N N - z 00 m W3 " 3 m m m - Nm w w 0 dW 0 v , P - O O W N N m m V I P - 00m - W m d o 0 0 0- m m 3 w m 0 N m 14 0 m m o o m N 0 N mW W N N m Om m w o m m d N E:d v , P - m m W d I . - I 5 s m c - 0 0 n o 0 - W N m *. 8 00 m r - a. -"op fl 7 3 N 0 0 -? 0 m b C m 0 0 N 2 W m vi 7 2 3 m O N z P- N 0 4' g i 3 2 N ~ . m 00 0 m00 2 P- N 0 o w r;' w s 2 3 0 d d P - a . 00 N m 0 s1 0 N f f r;' z v : 2 2 3 m 0 2 VI W P- D 0 vi N00 W m P- - c;' 2 D * 00 m w N 00 N m 2 0 m h E: N r;' 2 t 0 8 d N - e 00 N 7 d W i r;' -cP I 0 - vi m W P - N 00 Y 00 7 F h 2 0 0 3 N m I. 2 2 m m 0 0 2 3 op h 0 0 2 zm 2 2 W a. v: 00 W . -Y 3 N P 9 h 9 9 9 'E*B 3 4 * U -.-d P v 8 0 s -3 -E5 2 - 'SM 20e! 0 u . w iYc! .- m 3 3 c1 3 3 Y P 3 3 -I 0o + s i I C C C -r 0 C N 0 8 C N 0 N 0 0 N 0 54 3 C 0 N 0 0 U 0 -2 0 0 N0 D 9 C 9 0 9 C 9 C \ C > \ C ? \ . m id 3 .-8 i 26 a E . " m c j -r z PI yr) 0 0 N N 0 0 0 O b 0 N 3 N L 0 0 N 0 0 PI 0 m m Nm N 0 0 m 3 2 \o m z2- 0N N 0 0 m 9 . h 2 0 2 n 2 g i z i 8 0 N s3 0 W m N 0 0 N 0 3 m 0 0 d N m? 3 2 0 0 O O Q I 0 W N ", m 3 eI 91 9 91 00 QI 9 QI t- 9 W 91 9 3 3 e I vr 91 9 b, .-8 :: W Sff Z & C CC c c) 0 C 0 E N t r N 0 C 0 r1 N O c . I O N O r n 4 - m 0 W d m m I- h 0 0 0 s 0 * -m m m I0- 0 0 s t - - m mdN m W w o3 o ~ n-lns w n N o w m m 0 0 t - W - I- m s > 0 0 0 0 ~ 0 $ ! m W d W N ? - o w0m0m3o m n - m m dN N 00