01 — INTRODUCTION KEEP THE PACE NOVEMBER 2024 KEEP THE PACE How Inflation Erodes Cash Transfers And What to Do About It Ugo Gentilini, Hrishikesh TMM Iyengar and Giorgia Valleriani with Sheraz Aziz, Hana Rakhma Arimbi, Jusaine Lis Miranda Nogueira, María Angélica Trujillo, and Calvin Cheng 1 01 — INTRODUCTION KEEP THE PACE TABLE OF CONTENTS 1 / INTRODUCTION 6 2 / THE CASE FOR AND AGAINST (AUTOMATIC) INDEXATION 11 3 / RESULTS FROM GLOBAL INDEXATION DATABASE 15 Method of indexation 15 Institutional arrangements 25 Frequency and timing 26 4 / EVOLUTION AND ADAPTATIONS 30 From discretionary to automatic indexation 31 Argentina / 31 Australia’s social pensions / 33 Ghana: a real-time shift / 35 Change in method: from automatic to discretionary indexation 38 Maldives: from “conditional automation” to discretionary indexing / 38 No correction 39 Uruguay’s steady course / 39 Hybrid 41 Norway: indexation plus discretionary top-up/ 41 Mexico’s “unspecified automatic” model/ 42 Change in frequency (within automatic indexation) 45 Italy: from more to less frequent indexation / 45 Canada: from less to more frequent indexation /48 Change in mechanism: from prices to wages 49 New Zealand / 49 Mixing wages and prices 52 Germany’s supplementary indexation / 52 Change within prices 52 Belgium’s health index / 52 5 / STYLIZED FRAMEWORK 54 6 / CONCLUSIONS 59 REFERENCES 62 ANEXX 1 / TAXONOMY OF INDICES 76 ANEXX 2 / PROGRAM LEVEL INFO ON KEY FEATURES OF BENEFIT ADJUSTMENTS 77 2 >> KEEP THE PACE ACKNOWLEDGMENTS This report was made possible by the generous support of the World Bank’s Rapid Social Re- sponse Adaptive and Dynamic Social Protection (RSR-ADSP) Umbrella Trust Fund Program, which is supported by the Russian Federation, United Kingdom, Norway, Sweden, Australia, Denmark, the Bill and Melinda Gates Foundation, USAID, GHR Foundation and UBS Optimus Foundation. Sincere thanks go to several colleagues for their support throughout the report’s production pro- cess. The team is grateful to Jamele Rigolini and Loli Arribas-Banos for their strategic guid- ance; Josefina Posadas, Ruslan Yemtsov, and Zaineb Majoka provided precious peer review comments; Alessandra Marini, Clemente Avila, Cornelia Tesliuc, Dewen Wang, Federica Rical- di, Francesco Cenedese, Kenia Parsons, Ludivi- ca Cherchi, Manuel Salazar, Mohamed Almenfi, Nahla Zeitoun, Paolo Belli, Tiago Falcao Silva and William Wiseman were exceptional in sup- plying, validating or clarifying information on select countries; and the reflections by Colin Andrews, Marco Knowles and Herwig Immervoll shared at a seminar held at the World Bank were much appreciated. Finally, special thanks go to Nick Barr at LSE for early discussions on ideas developed in the paper. 3 >> KEEP THE PACE 01 — KEEP THE PACE EXECUTIVE SUMMARY The indexation of benefits, or anchoring cash transfers to inflation levels, rep- resents a key and underexplored dimension of the adaptive social protection (ASP) agenda. While considerable attention has been paid to coverage ex- pansion as a core function of ASP, this report argues that indexation can be fruitfully framed as a novel feature of making social protection systems more adaptive. Through indexation, the adequacy of cash transfers can evolve – or “keep the pace” – with changing conditions. This report applies an ASP framework to support policymakers in navigating trade-offs in indexation, including presenting new data and experiences to inform whether and how indexation could be calibrated in different contexts. The adjustment of cash transfers to inflation is more prevalent than often assumed, but it is often discretionary. This report offers a novel stockta- king comprising of 232 non-contributory cash transfer programs across 158 countries. These programs, which encompass unconditional cash transfers, conditional cash transfers, public works, and social pensions, are tracked using 16 indicators for a total of 7,056 datapoints. Almost four-fifth of the surveyed programs have some form of discretionary or automatic indexation, with about one-third of them doing so through automatic adjustments. Countries have dynamically evolved their approach to indexation significant- ly. The report’s 14 deep dives into specific country practices document that indexation practices have also evolved remarkably over time, including in terms of altering methods, mechanisms, and frequency of indexation. While indexation is nearly a standard feature in higher-income contexts, a rich set of experiences is emerging across the income spectrum, including salient real-time developments in lower income contexts. Different types of indexation present comparative strengths and limitations. A system that adjusts transfers discretionarily may have more control over fiscal costs; but it also places those decisions on potentially less predict- able and objective – indeed discretionary – decision making processes. The 4 >> KEEP THE PACE politics of transfer augmentation is greatly reduced, but not eliminated, by automatic indexation; the predictability of automatic benefits yields sizable benefits, but the mechanics of constructing indexation measures also rais- es a set of data and technical challenges. In cases of skyrocketing inflation, the balance between maintaining purchasing power and fiscal sustainability should also be carefully pondered. In fact, the appropriateness of discretion- ary, automatic and hybrid indexation modalities vary by context, with the level of maturity in ASP systems and prevailing rate of inflation shaping their via- bility, effectiveness and efficiency significantly. A rich operational agenda lies ahead. This includes tailoring the overall pa- rameters for indexation (whether automatic or discretionary), the appropriate selection of benchmark mechanisms between price, wage, or combinations thereof, and the customization of indexation to specific cash transfer de- signs. Furthermore, ironing out the thresholds and conditions under which indexation mechanisms, methods and parameters should change represents an important area of innovation. The practices and case studies distilling in this report represent an initial step in such direction. 5 >> KEEP THE PACE 1 / INTRODUCTION Food prices are growing steadily. Around mid-2024, at least ten countries displayed an annual nominal food inflation of over 40%, and in three of them it exceeded 100%. In Argentina, the rate was 293%. And up to two-thirds of middle-income countries had inflation higher than 5% (World Bank 2024). While inflation can be attributed to structural and idiosyncratic forces, long- term time series show that food prices, as measured by the Food Price In- dex1, have been climbing steadily over the past quarter century: real food prices in 2024 are about one-fifth higher than the previous decade and dou- ble the level of the preceding two decades (figure 1). Figure 1 / Trends in Food Price Index, 1990-2024 (2014-16 = 100) 140 120 100 80 60 40 20 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 Nominal Real Source: FAOSTAT, accessed September 2024 Linear (Real) Inflation is generally detrimental for economies and societies, although some may benefit from it. In the case of food inflation, a number of factors can affect the direction and depth of impact across society. For instance, in the short-term, net food consumers would be hurt by price increases. Such impact would hinge on the level of trade and market integration, 1 | The FAO Food Price Index consists of the average of five commodity group price indices (cereals, vegetable oil, dairy, meat, and sugar) weighted by the average export shares of each of group over 2014-2016 (see https://www.fao.org/ worldfoodsituation/foodpricesindex/en/) 6 01 — INTRODUCTION KEEP THE PACE which affect the degree of price signals transmission between nations and within countries; the market structure and elasticity of supply response would matter too, including the cost and time of adjustment to changing price signals by various actors along supply chains. In the medium-term, however, it is possible that higher food prices would incentivize agricul- tural, farm, off-farm and non-farm rural sectors for more market produc- tion and transactions. An augmented food supply may, in turn, reduce prices for consumers (of course, assuming that there are no hoarding or speculative practices by producers, wholesalers, intermediaries or retail- ers). Moreover, those price dynamics would be affected by what happens elsewhere in the economy, such as in labor markets and social protec- tion – both of which can affect effective demand and food entitlements2. The inflation debate encompasses broader fiscal policy issues. In infla- tionary settings, understanding the relationship between monetary and fiscal interactions is key. In some cases, the lines between the two are blur- ring: for example, the use of one-off injections of universal cash transfers in cases of low interest rates and demand contractions has been dubbed “unconventional” monetary policy3. As central banks consider how to ad- dress high inflation levels (via monetary policy) and governments strive to mitigate their negative welfare effects (via fiscal policy), it is important to ensure that the two levers operate within a coherent policy framework4. High food prices can have regressive effects. With the poorest sections of the population devoting higher shares of their budgets for food expendi- tures, high food prices can have catastrophic consequences. The right- hand of figure 2 shows such negative income-food expenditures relation- ship – the “Engel Law” – for Rwanda. The Law is likely to impact human capital via less nutrition and potentially reduced investments in human capital, both of which can have detrimental effects for long-term welfare. This effect is captured in the right hand of figure 2. 2 — See the classic work by Sen (1981), Timmer et al (1983) and Devereux (1988). 3 — See Gentilini (2022) 4 — See for example Gita Gopinath’s introductory remarks for the Conference “Fiscal Policy in an Era of High Debt”, November 17, 2023: https://www.imf.org/en/News/Articles/2023/11/17/sp-fdmd-gopinath-remarks-at-fiscal-forum- era-of-high-debt 7 01 — INTRODUCTION KEEP THE PACE Figure 2 / Share of household food expenditures out of total income Food budget share 10 8 6 4 2 10 12 14 16 18 log total expenditure Source: Nsabimana et al (2020) The relationship between cash transfers and inflation is debated empir- ically. Cash transfers are used to meet high price-induced needs: for in- stance, in 2023 at least 145 million people were reached by inflation-re- lated cash transfers in 17 countries5. While evidence generally suggests that cash transfers themselves don’t drive prices up, this can be the case under specific where market structures6. Such risk of augmenting infla- tion – or at least not being able to keep up with it – can lead people to prefer in-kind assistance, such as reported in India, Egypt and Ethiopia7 At the heart of the debate lies the definition and measurement of mar- ket functionality. Whether cash transfer may amplify inflationary effects hinges on market structure: where the elasticity of supply respons- es is low – because of structural bottlenecks interfering price signals (e.g., limited transport infrastructure) or market actors’ strategic be- havior – a significant injection of cash may result in higher prices. Con- versely, in contexts with an average degree of market functionality in place, cash transfers are unlikely to generate major, sustained surge in prices because of supply responses to increased effective demand8. Inflation moves the definition and calibration of transfers’ “adequacy” 5 — Gentilini et al (2023) 6 — Gentilini (2024), Allen and Gentilini (forthcoming) 7 — See for example Gentilini (2023). 8 — For a discussion on the effects of cash transfers on inflation, see Allen et al (2024) and Gentilini (2024) “Balancing Act: Navigating the In-Kind vs Cash Transfers Debate”, keynote speech at the Institute of National Planning, Cairo (recordings available at https://www.facebook.com/share/v/GxqXHoZ6WavkZQbv/?mibextid=WC7FNe). South Africa (Allison and Pillay 2024: decrease by design (for child grant, not other grants, and possible items) 8 01 — INTRODUCTION KEEP THE PACE centerstage. Trade-offs between coverage, adequacy, costs, and incen- tives have been recognized. As countries embark on a journey towards en- hanced social protection provisions, the travel can be filled with difficult choices and challenges. Policymakers situate choices around adequacy within broader considerations on coverage, incidence, incentives, fiscal costs, and political economy. Specifically, what constitutes an “adequate” transfer, that is, how much should beneficiaries ideally receive? Overall, transfers hinge on program objectives: for example, if the goal is to ensure a “healthy diet”, the average cost of it is estimated as $3.2-$3.76/day; some programs estimate the cost of accessing “minimum expenditure baskets”; programs can also anchor transfers to the poverty line (e.g., providing an equivalent of 20% of it) or to the minimum or prevailing wage (in the case of public works, for instance); transfers can also differ if a program is framed as offsetting opportunity costs (e.g., avoiding child labor), if it accounts for transaction costs incurred by beneficiaries (e.g., transportation), or if meant to simply offer a “reimbursement” (like in the case of some train- ings). Furthermore, if a scheme has goals other than food security – for example, related to assets (like livestock), rental assistance or adherence established legal rights – the corresponding amount can be commensu- rate to those (likely higher) spending rationales. There might be a difference between planned and actual amounts of trans- fers that people receive. After a program’s objective has been defined and transfers calibrated accordingly, a range of circumstances can lead to lower provisions (figure 3). For instance, societal and political attitudes towards the notion of “deservingness” may prevent some people in need to receiving cash assistance (e.g., working age populations in the labor market earning low wages). Funding shortfalls may allow to cover a subset of eligible par- ticipants and place others in waiting lists. Delivery bottlenecks, like limited reach of ID or payment systems, and high transaction costs for beneficiaries, such as travel time or excessive documentation requirements, can further stymie participation. People may incur in costs for accessing benefits, like for transportation. And finally, the purchasing power of cash transfers can be eroded and even wiped out by inflation. The difference between planned and actual transfers can have profound effects on program performance. 9 01 — INTRODUCTION KEEP THE PACE Figure 3 / Stylized factors hindering cash transfers design Objetives Societal attitudes Funding Delivery Beneficiary transaction costs Purchasing power erosion Source: authors, adapted from OECD (2024) Indexation can be interpreted as an act of realignment (“keeping the pace”) between inflation and transfer size. The former changes continuously, while the latter is often adjusted more sporadically and unpredictably, if at all. While some components of social protection systems have been an- chored to inflation: for example, there is a century-long tradition with con- tributory pensions indexation, with Denmark’s seminal measures in 1933. Yet, indexation in social assistance has been less prevalent. The core challenge tackled by the report is to understand whether to consider cash transfers indexation, the practical ways to do it, how these may evolve, and the trade-offs involved in the indexation process. The report helps filling a longstanding gap by offering a novel stocktak- ing comprising 232 non-contributory cash transfer programs across 158 countries. These programs, which encompass unconditional cash transfers, conditional cash transfers, public works, and social pen- sions, are tracked using 12 indicators for a total of 7,056 datapoints. A full excel database accompanies this report, alongside a set of 14 short case studies providing more granularity into country practices9. The report is structured as follows. The next section sketches out the ben- efits and limitations of indexation; section 3 offers an overview of findings from our global stocktaking; section 4 illustrates select program trajecto- ries in indexation choices over time; section 5 sets out a stylized frame- work for indexation choices, while section 6 concludes. 9 —These include Argentina, Australia, Belgium, Brazil, Canada, Chile, Germany, India, Italy, Maldives, Mexico, New Zealand, Norway and Uruguay. 10 02 — THE CASE FOR AND AGAINST AUTOMATIC INDEXATION KEEP THE PACE 2 / THE CASE FOR AND AGAINST (AUTOMATIC) INDEXATION Indexation to benefits can be undertaken on a discretionary or automa- tic basis. In the former case, policy makers undertake ad-hoc legislative action that defines the modality and level of adjustment. In addition, the government can also decide whether and when to make such discretion- ary adjustments. In the case of automatic adjustments, benefits are upra- ted based on planned, predefined rules (e.g., benchmark mechanism) and frequency. By providing regular and transparent adjustments, uncertainty surrounding the benefit amount – and losses of purchasing power – are likely reduced. Yet both options present technical and political economy advantages and drawbacks. Anchoring benefits on predefined triggers that adjust the size of cash transfers automatically involves complex considerations. The case for au- tomatic indexation rests on maintaining program effectiveness, coordina- tion with contributory arms of the social protection system, the reduction in “hidden costs” born by beneficiaries when benefits are not adjusted, and a set of qualitative considerations around transparency and predictability. Drawbacks may include the possible explicit high cost in high inflation- ary contexts, technical difficulties related to calibrating and benchmarking benefits, and, from a policymaker perspective, perhaps less room for (but not absence of) flexible and idiosyncratic changes. Inflation can have a corrosive effect on the ability of cash transfers to at- tain their stated goals. Some examples can illustrate the latter point. In Brazil, the size of cash transfers under the flagship Bolsa Familia pro- gram has increased steadily over time. While Brazil has no automatic ad- justment, between 2003 and 2020 benefits were increased seven times discretionarily. For a family of two adults and two children living in ex- treme poverty, nominal benefit values doubled from $16 to $34 (figure 4, left graph).10 The same graph indicates a widening gulf between growth 10 — During the COVID-19 period (2020-21), the government had implemented two additional programs, Auxílio Emergencial and Auxílio Brasil. And immediately after the pandemic, the government introduced a new program (called New Bolsa Familia). The benefit amounts in these schemes were considerably higher than those under the traditional Bolsa Familia. Some estimates suggest that post pandemic benefit levels have been almost 300% higher than pre-pan- demic transfers. 11 02 — THE CASE FOR AND AGAINST AUTOMATIC INDEXATION KEEP THE PACE in the minimum wage and Bolsa benefits. Considering the general in- flation index (INPC) and food inflation index (INPC-food), and the min- imum wage (MTE12_SALMIN12), in 2021 the real value of Bolsa benefits was reduced by about 22%, 33%, and 60%, respectively (Figure 4, right graph). Similarly, there has been a widening gulf between actual and hy- pothetically indexed benefits in Chile’s Ethical Family Income program: such gap peaked in 2020, when the difference amounted to around 23%11. Figure 4 / Gap between actual and inflation-indexed benefits in Brazil 90 Average benefit ($) 85,7 10 % gap of actual benefits vs inflation-adjusted benefits 80 70 60 -15 -11 50 43,8 40 34,2 -33 30 -40 20 10 -60 0 -65 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021 Value received % gap Benefit (INPC) Value adjusted with inflation (INPC) % gap Benefit (INPC FOOD) Value adjusted for min wage Value adjusted for min wage Source: authors based on data from various government official materials (see references) In other cases, an automatic indexation was incorporated in program de- sign and acted as an embedded inflation shield. India’s Mahatma Gandhi National Rural Employment Guarantee Act (NREGA) is a case in point. The scheme is a large-scale public works program that covered about 61 mil- lion people in 2022-23. When introduced in 2011, NREGA was anchored on linked to the Consumer Price Index for Agricultural Labour (“CPI-AL”). The periodic revision of wages is automatic and happens at the beginning of each financial year in April based on CPI-AL of December of the previous year. Estimates produced for this report show that in the absence of index- ation, by 2023 the average wage would have lost 37.9% of its purchasing power of a decade earlier (2013/2014) (figure 5). 11 — The program, also known as Securities and Opportunities, was introduced in 2012 and provides cash transfers to about 3.3% of the population living in extreme poverty. 12 02 — THE CASE FOR AND AGAINST AUTOMATIC INDEXATION KEEP THE PACE Figure 5 / Loss in purchasing power in India’s NREGA in the absence of wage indexation (base = 2013/2014) Loss of purchasing power in the absence of benefit adjustment -2,46% 0% -5,20% -11,63% -10% -13,31% -17,33% -20,80% -20% -28,83% -30,55% -30% -34,23% -37,91% -40% 2014/2015 2015/2015 2016/2017 2017/2018 2018/2019 2019/2020 2020/2021 2021/2022 2022/2023 2023/2024 Source: authors based on data from various government official materials (see references) Inflation can be interpreted as a tax on benefits. If not indexed, such tax may be borne entirely by beneficiaries as opposed to being shared by the larger societal pool (via potentially increased funding for benefit ad- justment). The case of Iran in 2011, for instance, shows how absent mea- sures for updating benefit rates, can wipe out almost the entire purchasing power of transfers and represent an “exit strategy” for such provisions12. The benefits of a transparent, accountable, and predictable system of in- dexation can generate beneficial outcomes beyond technical parameters. One of them might be the notion of the state being responsive to evolving economic conditions. Another relates to mitigating price shocks ex-ante as opposed to acting via discretionary, ex-post increases. In other words, indexation reduces uncertainty and volatility, thereby likely generating en- hanced psychological outcomes. This makes indexation a key (and under- explored) component of adaptive social protection systems in general, and of anticipatory cash transfers in particular. Yet no indexation mechanism is perfect, and real-world evolving conditions may not be accurately re- flected in indexation systems. Basis risk exists in insurance as well as in indexation. There are arguments made against indexation. One of them is the fiscal burden. The additional costs required to keep up with inflation can be non-trivial. However, the extent to which such injection is significant de- pends on program goals and duration (e.g., social pension versus, say, a 12— See Gentilini et al. (2020) 13 02 — THE CASE FOR AND AGAINST AUTOMATIC INDEXATION KEEP THE PACE guaranteed minimum income scheme), design (i.e., the share of transfers out of total operational costs), the magnitude of the additional transfer, and current level of spending. Estimates presented in this report show that the additional costs for indexation in most cases may not be prohibitive. In the recent case of Ghana, for instance, the doubling of benefits of the LEAP program would only generate a 40% increase in cost. The fact that indexation makes a system more objective implies a reduced role for discretionary injections. This may reduce flexibility and agency by policymakers. The main point is that cash transfers can also be interpreted with a wider political lens whereby the provision and design of programs can generate electoral returns13. Within such framework, the adequacy of transfers can offer a discretionary option to tap to build or cement politi- cal consensus. While the issue would deserve further empirical attention, such a political economy perspective may constitute an important dimen- sion in preserving and expanding transfer size in contexts of competitive politics as well as productivists regimes14. From this standpoint, it might not be unreasonable to expect some resistance in introducing objective, automatic and trigger-based systems that are not associated with a par- ticular political party or incumbent. Yet, as this report shows, discretionary and automatic indexation are not mutually exclusive. There can be a role for policy discretion, chiefly in terms of top-ups, even within an automatic systems of benefit adjustment (e.g., New Zealand). 13 — a growing body of literature is investigating the effects of conditional cash transfers on electoral politics, especially in Latin America and East Asia. In general, a positive effect is found in support of incumbent parties (https://www.jour- nals.uchicago.edu/doi/10.1086/701211). There is also emerging evidence of similar effects in high-income countries like Poland (https://onlinelibrary.wiley.com/doi/full/10.1111/ecca.12505) 14 — See Hickey et al (2019) 14 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE 3 / RESULTS FROM GLOBAL INDEXATION DATABASE This section provides an overview of findings organized along three key areas for “keeping the pace” with inflation: (i) the method of indexation, i.e., automatic or ad-hoc adjustment – or even no adjustment at all; (ii) bench- mark mechanisms or reference, i.e., changes in prices, wages, macro- economic variables, etc.; and (iii) other parameters to anchor indexation decisions, including frequency and timing of benefit changes (i.e., annual, semi-annual, quarterly, monthly, etc.). METHOD OF INDEXATION Benefit adjustments of cash-based transfers can either be automatic or implemented on an ad hoc basis. Automatic adjustments are generally based on a statutory obligation to synchronize benefit amounts to chan- ges in a specific benchmark mechanism (such as price indexes, wages) based on a predetermined frequency and timing. In contrast, for ad-hoc adjustments, governments decide methods, benchmark, frequency, and timing in ways that are relatively less automated or more discretionary and idiosyncratic. Most programs present some form of adjustment, especially ad hoc. Around 79% of programs (183 out of 232) across 126 countries adjust be- nefits, while programs with either no adjustment (26 out of 232) or no in- formation available (23 programs) claim about 10% of the sample each (see figure 6). Among the programs that have adjusted benefit amounts, the majority (68%; 124 programs across 97 countries) perform ad hoc changes and about one third (32%; 59 programs across 38 countries) display an au- tomatic adjustment. 15 —RESULTS 03 / FROMGLOBAL RESULTSFROM INDEXATIONDATABASE GLOBALINDEXATION DATABASE KEEP THE PACE Figure 6 / Indexation methods for cash transfer programs (n= 232 programs; n= 158 countries) 250 No. of programs 200 232 150 100 124 50 59 26 23 0 Automatic Adhoc (N=97) No (N=20) No data (N=23) Total sample (N=39) (N=158) Source: authors based on indexation database for this report. Note: N= No. of countries; n= No. of programs. “No” represents programs whose benefit size has not changed over time. “No data” represents programs with no information available on adjustment type. The use of automatic benefit indexation tends to increase with country in- come level. Out of the total sample of automatically indexed programs, 93% stem from high income and upper middle-income countries. Only 4 programs are present in low and lower-middle income contexts, namely Ghana (Liveli- hood Empowerment Against Poverty, LEAP), India (Mahatma Gandhi National Rural Employment Guarantee Act, MGNREGA), Malawi (Food and Cash Trans- fers program, FACT), and Tajikistan (Targeted Social Assistance, TSA). Over half (53%) of indexed cash transfers hail from Europe and Central Asia (31 programs, 53%), followed by Latin America and the Caribbean, and East Asia and Pacific with 12 programs (20%) and 6 programs (10%), respectively (see figure 7). Only 7 automatic programs are available across the Middle East and North Africa, Sub-Saharan Africa, and South Asia. 16 —RESULTS 03 / FROMGLOBAL RESULTSFROM INDEXATIONDATABASE GLOBALINDEXATION DATABASE KEEP THE PACE Figure 7 / Distribution of program indexation methods, by region and country-income group (n= 183 programs; n= 126 countries) 140 No. of programs 124 120 100 80 59 60 44 38 40 40 31 33 31 28 26 21 15 20 12 9 8 8 6 2 2 3 1 3 4 1 0 SAR MENA NORTH AFR EAP ECA LAC GLOBAL LIC LMIC UMIC HIC Authomatic (N=7) (N=9) AMERICA (N=23) (N=17) (N=43) (N=26) (N=127) (N=9) (N=38) (N=40) (N=40) Adhoc (N=2) Source: authors based on indexation database for this report. Note: N= No. of countries; n= no. of programs (183 programs = 124 ad hoc + 59 automatic). SAR = South Asia Region, MENA = Middle East and North Africa, AFR = Sub Saharan Africa, EAP = East Asia and Pacific, ECA = Europe and Central Asia, and LAC = Latin America and Caribbean. LIC = Low Income Countries, LMIC = Lower Middle Income Countries, UMIC = Upper Middle Income Countries, HIC = High Income Countries. Among different program types, social pensions are the most automatical- ly indexed form of cash-based transfers. Out of a total of 56 social pension programs, 38% (or 21 programs) display automatic indexation. Uncondi- tional cash transfers, however, feature the highest number of automatic programs (29 out of 106 programs), which represent a share of 27%. About 13% and 10% of conditional cash transfers and public works programs have automatic adjustments, respectively (figure 8). Figure 8 / Indexation methods by cash transfer instrument (n=232 programs) Composition of programs with indexation 0% 25% 50% 75% 100% Social Pensions 38% 55% 5% 2% (n=56) Unconditional Cash Transfer 27% 54% 8% 10% (n=106) Conditional Cash Transfer 13% 57% 22% 8% (n=60) Authomatic Public Works Adhoc (n=60) 10% 20% 10% 60% No No Data Source: authors based on indexation database for this report. Note: n= No. of programs. 4 programs used a mixed instrument approach, 2 of them used UCT and CCT, and the other two used UCT and Public Works Programs. Based on program description and the relative importance of those components, we classified the first two programs as CCTs and the other two as UCTs. 17 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE A total of 26 programs, or 11% of the sample, have not adjusted benefits over the years. In Bolivia, programs like Juancito Pinto and Juana Azur- duy de Padilla have maintained their benefit amounts unaltered. In both instances, top-ups (worth of US$72 and US$58, respectively) were intro- duced in response to the Covid-19 pandemic. Similarly, Guatemala’s Social Food Basket, Thailand’s Welfare Card scheme, Zimbabwe’s Harmonised Social Cash Transfer, Burkina Faso’s Nahouri Cash Transfers Pilot Project, Republic of Congo’s Lusingi project, and Burundi’s Cash for Jobs project have all not adjusted their benefits over the years. In some cases, indexation was established formally, but not consistently applied in practice. In 2016, Tanzania’s Zanzibar universal pension pegged benefit rates to the CPI, albeit there is anecdotal evidence that benefits ha- ven’t been raised consistently. Mexico’s Prospera used to automatically adjust transfers (following biannual review) based on the national CPI of the basic basket published by the Bank of Mexico. In 2012, the CPI was replaced by an index of urban and rural Minimum Welfare Lines published by CONEVAL (capturing real value of a monthly food basket per person). Nonetheless, between 2014 and 2019, the transfer value did not change un- til the program was withdrawn. Finally, Mozambique’ Basic Social Subsidy Programme (PBSS)15 has not adjusted its benefits since 2018. As per the recommendation by the National Basic Social Security Strategy (ENSSB-II; 2016-24), the value of the transfers should have been annually adjusted to inflation. Had the benefits been properly revised, as of 2021 a single-per- son household should be receiving MZN 650 instead of the MZN 54016. There are also instances of indexation practices being introduced and subsequently withdrawn for fiscal constraints. Finland’s child Benefit (Lapsilisä) was supposed to be annually adjusted to the National Pen- sion Index17 starting from March 2011. The first increase of 0.4% was im- plemented for 2011, and a subsequent adjustment of 3.8% was also intro- duced for 2012. But in 2013, Section 21 of the Child Benefit Act (713/2012) was amended to temporarily suspend the indexation for the 2013-15 pe- riod. The Act envisioned to reinstate indexation starting from 2016. How- ever, a further amendment to the Child Benefit Act (1661/2015) resulted in abolishment of indexation of child benefits. In tandem with reductions 15 — An unconditional cash transfer – which is one of the four main social protection responses in the country. Under PSSB, the monthly values for benefits are: (a) MZN 540 for one-person households; (b) MZN 640 for two-person house- holds; (c) MZN 740 for three-person households; (d) MZN 840 for four-person households; (e) MZN 1,000 for five-persons households (ILO & UNICEF, 2021). 16 — ILO and UNICE (2021) 17 — This was linked to the changes in the consumer prices. 18 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE in benefits levels, the demise of indexation reflected fiscal pressures18. INDEXATION MECHANISMS Most of the automatically adjusted cash programs are indexed on prices. Information on benchmark mechanisms for benefit adjustment is available for 109 programs, of which 50 are ad hoc adjusted programs while 59 are automatic. Programs across the globe use different forms of benchmark mechanisms, such as changes in prices, wages, macroeconomic variables (e.g., tax collection), or some combination thereof (see figure 9). Adjust- ing benefit amounts based on prices constitutes the most popular mech- anism, accounting for more than half of the programs (i.e., 52%, or 57 pro- grams out of 109). The large majority of price-based indexation features automatic adjustments (75%), while most ad hoc programs involve a “dis- cretionary” mechanism. This applies to 29 programs that may not always have a fixed or pre-determined benchmark, and instead hinge upon factors like budget availability, political economy (e.g., sudden increase in benefits around elections), and other discretionary considerations. Figure 9 / Benchmark mechanism for benefit adjustment (n= 109 programs) Prices 43 14 Wages 7 4 Wages and 3 Macro-economic variables Prices, Wages 1 and other factors Prices and 1 Macro-economic variables Discretionary 29 Prices, Wages and Macro-eco- nomic variables 2 0 10 20 30 40 50 Authomatic Adhoc No. of programs Source: authors based on indexation database for this report. Notes: n= no. of programs. Based on 109 programs, 59 have Automatic adjustments while 50 have Ad hoc adjustments. 18 — For instance, in 2015 the child benefit levels (expect for single parent supplement) were reduced by 8.1% (Act amend- ing Section 7 of the Child Benefit Act, 1111/2014). Similarly, in 2017 the benefit levels (except single parent supplement) were reduced by 0.91%. At the time of proposing the motion, reducing the level of child benefit was estimated to save 11.7 million euros in child benefit expenses in 2017 and 47 million by 2020. For more information, see https://www.finlex.fi/fi/ esitykset/he/2016/20160151#idm46111192815824 19 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE Specifically, there is a clear tendency towards indexation anchored on Consumer Price Index (CPI). Among the 66 programs using price bench- marks, 32 programs (or 48%) used CPI for all goods (figure 10), while the rest used some variations of CPI. As discussed, in India wages of MGN- REGA public works program have been linked to CPI-AL since 2011.19 By contrast, Italy’s benefits of the Incapacity Pension are linked to the index of consumer prices of blue- and white-collar families (FOI), which refers to the consumption of families headed by an employee from the non-agri- cultural sector (see section 4 for more details). Similarly, the United States’ Supplemental Security Income (SSI) program anchors its benefits to the CPI for Urban Wage Earners and Clerical Workers (CPI-W). Another inter- esting variation of CPI comes from Belgium, where benefits of several so- cial assistance programs, such as the Guaranteed Minimum Income Ben- efit for the Elderly (GRAPA) and Child Benefits, are anchored to an index that excludes products such as tobacco, fuel and alcoholic beverages, i.e., the Smoothed Health Index. Similarly, in France, several social benefits, in- cluding the Active Solidarity Income (RSA) programs, are indexed on CPI excluding tobacco. 19 — The CPI-AL basket and weights are based on consumption expenditure data collected during the National Sample Survey 38th Round of Consumer Expenditure Survey in 1983. In such survey, agricultural labor households were defined as households with at least 50% of total income coming from manual labor in agriculture in the previous year. See also discussion in the case studies section. 20 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE Figure 10 / Types of price-based benchmark mechanisms (n= 66 programs) CPI 26 6 CPI for Urban Wage Earners and Clerical Workers 1 (CPI-W) CPI for Agricultural Labour (CPI-AL) 1 Consumer prices for blue and white-collar families 1 (FOI) CPI of low-income households 1 CPI and Harmonised Index of Consumer Prices (HICP) 1 CPI and sub-indices cost of rented/owned housing 1 CPI and adjustable reference unit (set by decree) 1 CPI excl. tobacco 1 Smoothed Health index 2 Minimum subsistance (calculated based on 3 inflation) Reference Social Indicator (RSI) [indexed to 1 inflation/prices] Avergae price development at federal level 1 Average price inflation 1 Market food prices 1 Locally available nutritionally-balanced food basket 1 Basic food basket 1 Not specified 6 8 0 10 20 30 Authomatic Adhoc No. of programs Source: authors based on indexation database for this report. Notes: n= No of programs. Based on 66 programs, 49 have Automatic adjustments while 17 have Ad hoc adjustments. N.B., there is a small overlap among Figures 10, 11 and 12 due to mixed adjustment mecha- nism used by the programs (e.g., price and wages; prices, wages and macro-economic variables; etc.). 21 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE Price indexation can go beyond CPI. For Argentina’s Citizen Program (Pro- grama de Ciudadanía Porteña), the update of the benefit is carried out ev- ery six months based on the variation of the basic food basket, which also determines the country’s poverty line as estimated by the National Insti- tute of Statistics and Census of Argentina (INDEC). In Ethiopia, the PSNP20 scheme reviews public works wages based on the annual food price CPI published by the Central Statistical Agency. Adjustments to Estonia’s so- cial pension (minimum pension guarantee) are based on two components: CPI and a social tax revenue (a levy on income to finance pensions and state health insurance). While in the past those components had an equal weight in estimating changes to benefits, the current approach accords a weight of 80% to the social tax and 20% to CPI. Some ad hoc programs seemingly behave as automatic adjustments. The UK’s Universal Credit is a case in point: the Social Security Administration Act of 1992 requires the Secretary of State to annually review benefits and pensions. However, by convention Universal Credit is uprated in line with increases in the CPI every September. In the context of wage indexation, programs are typically anchored on av- erage wages. In Denmark, benefits of the non-contributory Public Pension scheme (Folkepension) are adjusted annually in line with average earn- ings. Similarly, in Uruguay, old-age and disability social pensions are re- vised annually based on changes in the average wage index (Índice Medio de Salarios) of the previous year. In the Netherlands, the Social Assistance scheme (Participatiewet) revisions in transfer values take place every six months and are linked to the statutory minimum wage, while the asset test limits are indexed annually based on the CPI. Another example comes from Brazil’s Continuous Benefit Program (BPC) where the benefits are adjusted annually based on changes in the legal monthly minimum wage, which is in turn indexed annually to a formula anchored on price inflation and GDP growth. Most wage-based mechanisms rely on statutory minimum wages. Such mechanism has been recorded for 4 programs, followed by average earn- ings or average wages (figure 11). The latter method could refer to average ordinary time weekly earnings, as in the case of New Zealand’s Sole Par- ent Support and Job Seeker Support Program, or specific wage indices, as for Barbados’ Old-Age Assistance Pension and Uruguay’s non-contributory 20 — Productive Safety Nets Program 22 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE pensions. Among programs that adjust on a discretionary basis, there are benefits anchored to minimum income (e.g., Latvia’s Guaranteed Minimum Income benefit),21 entry-level salaries (e.g., Libya’s Basic Pension Bene- fit) and wages in the labor market for the program implemented areas (e.g., Guinea’s Labor-Intensive Public Works Program). Figure 11 / Types of wage-based benchmark mechanisms (n= 22 programs) Variation of average taxable remuneration of 3 stable workers Average earnings 3 Net average ordinary time weekly earnings 2 Average wage index 2 Average earnings of rural and urban residents 1 Average development of the net wages and salaries 1 per employee Not specified 1 Wages in the labor market of the [program’s] interested areas 1 Median minimum income 1 First-grade starting salary 1 Statuory minimum monthly wage 3 1 Increase of wage of employees 2 0 1 2 3 4 Authomatic Adhoc No. of programs Source: authors based on indexation database for this report. Notes: n= No. of programs. Based on 22 programs, 15 have Automatic adjustments while 7 have Ad hoc adjustments. N.B., there is a small overlap among Figures 10, 11 and 12 due to mixed adjustment mechanism used by the programs (e.g., price and wages; prices, wages and macro-economic variables; etc.). Some programs have adopted mechanisms other than price and wage- based approaches. This includes a limited number of programs (7 in total, see figure 12). Argentina falls within this category, with social assistance 21 — In Latvia, minimum income is 20% of median equalized disposable income. 23 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE programs – such as, the Non-contributory Pension, Universal Pension for the Elderly (PUAM) and the Universal Child Allowance (AUH) – regu- larly adjusted based on a new “mobility rule.” This approach is based on the sum of 50% of the quarterly increase in the collection of the National Social Security Administration (ANSES) revenues and 50% for the salary variation of state employees (RIPTE index) of the same period, as calcu- lated by the National Institute of Statistics and Census (INDEC). Another mix-method practice is found in China, where the Old Age pension for rural and non-salaried urban residents are regularly adjusted based on the con- sumer price index, the average earnings of rural and urban residents, and the basic pension benefit levels of urban workers. Figure 12 / Mix benchmarks (n= 7 programs) Movility Index (evolution of salaries and tax 3 resources) Budget of the State, economic situation and increase of wages of employees 2 CPI, avergae earnings of rural and urban residents, 1 benefit amounts of other social security programs CPI and social tax revenues 1 0 1 2 3 4 Authomatic Adhoc No. of programs Source: authors based on indexation database for this report. Notes: n = No. of programs. Based on 7 programs, 5 have Automatic while 2 have Ad hoc Adjustments. N.B., there is a small overlap among Figures 10, 11 and 12 due to mixed adjustment mechanism used by the programs (e.g., price and wages; prices, wages and macro-economic variables; etc.). Top-ups to automatic indexation can also occur. In Mauritius, the benefits of the Basic Retirement Pension program are synchronized annually ac- cording to nominal wages. A top-up is provided on a 5-year basis, includ- ing a non-statutory correction to pension rates alongside wage indexation. In Estonia, the formal indexation methodology has been accompanied by discretionary increases implemented by different government coalitions. In New Zealand, the Jobseeker Support program has experienced three ad hoc adjustments over the past decade, and these were implemented in addition to (i.e., on top of) the automatic adjustments (called the Annu- al General Adjustment) that takes place in April of each year. Most recent example was in 2023, the cabinet approved a one-off increase in benefit amount by 0.98 percent22 on 1 April 2023 on top of the wage indexation, so 22 — 0.98 = CPI – Net average wage increase = 7.22% - 6.24%. 24 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE that the total increase in benefits was equivalent to the increase in the CPI (all groups), which was at 7.22 percent.23 Some programs index benefits to minimum subsistence estimates. At least 7 programs across countries such as Kazakhstan, Moldova, Russia, Romania, Slovakia, and Slovenia have such indexation in place. In par- ticular, government agencies calculate minimum subsistence income for each beneficiary unit, and that value is automatically indexed to changes in prices. For example, Slovakia’s Assistance for Material Needs program (Pomoc v hmotnej núdzi ) uses minimum subsistence income to determine the benefit amounts. The program is indexed each year on January 1 tying benefits to subsistence minimum estimated on July 1 of the previous year. In addition to the regular CPI, Slovakia calculates CPI for different social groups (e.g., low-income households and pensioners) by allocating differ- ent weight for individual items in the consumption basket based on the expenditure patterns and consumer behavior of a particular social group. INSTITUTIONAL ARRANGEMENTS Ad-hoc adjustment can involve a degree of unpredictability as stemming from institutional budgetary, and political decisions. For example, in Ar- menia’s Family Poverty Benefit transfers are adjusted on ad hoc basis de- pending on availability of funds. Latvia’s childcare benefit (Bērna kopšanas pabalsts) is adjusted based on fiscal space. Under Pakistan’s BISP, a com- mittee meets every year to review and decide whether to increase the ben- efit amount based on the fiscal envelop and many other factors including inflation. The Czech Republic’s Parental allowance is adjusted on discre- tionary decision of the government. Norway’s child benefits (kontantstøtte) adjust cash amounts yearly based on the annual national budget. Swe- den’s Ekonomiskt Bistånd/Försörjningsstöd (minimum income benefit) is not automatically indexed, with parliament deciding annually the rates updates (scale rates are set nationally and municipalities are allowed to top them up). And in Bulgaria, benefits are adjusted on discretionary basis, where the amount of the Guaranteed Minimum Income is determined by an Act of the Council of Ministers. 23 — This was part of the Welfare Assistance package to help low-income people meet the increasing cost of living. For the year to December 2022, CPI increased by 7.22 percent while net average wage growth was only 6.24 percent, result- ing in benefit amounts not rising enough to meet the increasing cost of living. So, the government approved to increase benefit amounts by an additional 0.98 percentage points, which translates into an increase of benefits between $2.29 and $6.26 per week (MSD, 2023a). 25 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE Certain indexation mechanisms may also be based on indicators that are tracked at a more local level, while in other countries formal indexation is often complemented by discretionary interventions. In Germany, for in- stance, the Citizen Benefit (Bürgergeld) is updated annually (on January 1) in two steps: in a first one (“basic update”), rates are tied to a mixed index, which is 70% based on the average price development at federal level, while 30 % is based on the average development of the net wages and salaries per employee. As a second step (“supplementary update”), the amount re- sulting from the basic update is further updated on the basis of most recent price development in the economy. In Austria, the value of the Minimum In- come (Bedarfsorientierte Mindestsicherung/Sozialhilfe) is updated annu- ally based on increases in the supplementary pension (Ausgleichszulage). As of January 2020, there is no national minimum standard, and benefits are regulated differently across states. A slightly different case is found in the United States: the Temporary Assistance for Needy Families (TANF) is a highly decentralized program, so in this case the autonomy to automatically index benefits to inflation is provided to state governments. Some of these states have not increased benefits (e.g., Pennsylvania) while other states have automatic adjustments (e.g., D.C., Connecticut). Inflation adjustments present particular challenges in humanitarian con- texts.24 In Yemen, the cost of Survival Minimum Expenditure Basket (SMEB) commodities are monitored for three consecutive months and a SMEB Technical Working Group meeting is conducted for deliberations once in every three months. In addition, an ad-hoc meeting is called whenever the cost of commodities breaches a threshold of 95% of the transfer value. In north-west Syria, partners and financial service providers (FSPs) contract exchange rates the day before or the same day of cash disbursements. Under the Lebanon One Unified Inter-Organizational System for E-cards (LOUISE), development partners were able to negotiate with the govern- ment a preferential exchange rate. In Malawi, under the FACT program pric- es were measured four times. Three commodities were monitored across seven markets where FACT was implemented, including prices for maize, beans and cooking oil. FREQUENCY AND TIMING Most automatically indexed programs adjust benefit amounts on an an- nual basis. Data on frequency of adjustment is available only for 121 pro- grams, of which 68 are ad hoc programs while 53 have automatic adjust- 24 — See ECHO (2022), CALP (2021) and McLean et al (2021). 26 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE ment. Almost half of programs perform annual adjustments, 60 of 121 programs (Figure 13). Programs that adjust on annual basis, mostly use automatic adjustments (i.e., 39 programs; 65% of 60 programs) and only 35% of the programs (21 of 60) perform ad hoc adjustments. The next most-used adjustment frequency is “discretionary,” representing over one third of the programs. This adjustment frequency only pertains to ad hoc adjusted programs as they might not have a fixed frequency for adjusting the benefits. Besides annual and discretionary, there are a range of other adjustment frequencies. On one end, we have higher adjustment frequen- cies (such as semi-annual, quarterly and monthly) which are mostly used in countries with high or very high inflationary situations. On the other end, we have programs with lower adjustment frequencies, such as adjust- ments once in two or three years, and adjustments when the CPI or wages breach a certain threshold. Figure 13 / Frequency of adjustments (n= 121 programs) Annual 39 21 DIscretionary 41 Others (e.g. one in two or three years) 5 Semi-anuual 4 1 Monthly 4 Quarterly 4 Growth of benchmark indicator above a specific 2 threshold 0 10 20 30 40 50 Authomatic Adhoc No. of programs Source: authors based on indexation database for this report. Notes: n = no. of programs. Based on 121 programs, 53 have Automatic while 68 have Ad hoc adjustments. The majority of programs with automatic indexation adjusts transfer amounts on an annual basis. Indexation is often undertaken in January, as in the case of the US’ Supplementary Security Income (SSI) scheme, Germany’s Citizen Benefit, Mauritius’ Basic Retirement Pension Scheme (BRP), and Mexico’s Pension for the Well-Being of Older People. In Italy, the Incapacity Pension rates are adjusted every January based on the provi- sional FOI index. A number of programs revise benefits in April – such as 27 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE India’s MGNREGA, Estonia’s Sotsiaalkindlustusamet, Moldova’s Ajutor So- cial and France’s RSA – while others are done in February (Chile’s Securi- ties and Opportunities/Ethical Family Income), March (Slovenia’s Financial Social Assistance), July (Canada’s Child Benefit and Slovakia’s Minimum Income Scheme), October (Ghana’s LEAP), and by localities with a large variation of time for China’s Dibao. A range of programs change benefits at other intervals. Four programs ad- just benefits on a quarterly basis. In Argentina, benefits of the Universal Pension for the Elderly (PUAM) and the Universal Child Allowance (AUH) fol- low the mobility rule, which update benefit in March, June, September, and December. In Canada, the Old Age Security (OAS) scheme payments are re- viewed each year in January, April, July and October, based on the difference between the average CPI of (i) the most recent 3-month period for which the CPI is available, and (ii) the last 3-month period where a CPI increase led to an increase in OAS benefit amounts. Some programs with available informa- tion adjust benefits bi-annually. This is the case of the Porteña Citizenship Program in Argentina, the Social Assistance program in the Netherlands and the Age Pension program in Australia, where the update of the benefit is car- ried out every six months in March and September. About two-thirds of programs with ad hoc adjustments don’t have a fixed frequency. Data on benchmark indicator for ad hoc adjustments are avail- able for 68 programs (out of 121; 56%). Of that, 41 programs (60%) do not use any specific frequency but adjust benefits on discretionary intervals. In cases where Ad hoc programs use a specific frequency, they most- ly perform annual adjustments (21 programs; 31%), followed by once in two or three years (5 programs, 7%) and remaining one program performs semi-annual adjustments. For example, Chile Solidario program adjusts benefits annually on an ad hoc basis, while Bolivia’s Renta Dignidad pro- gram25 benefits adjust once in every three years by the Executive Branch based on available funds. Next, it is interesting to note that the frequency of benefit adjustment may varies by program even within the same coun- try, as for Lesotho’s Old Age Pension (OAP) and Child Grant Program (CGP). The former has been steadily uprated, while the latter remained flat over prolonged periods of time (figure 14). 25 — Also known as Renta Universal de Vejez 28 03 — RESULTS FROM GLOBAL INDEXATION DATABASE KEEP THE PACE Figure 14 / Nominal benefits for Lesotho’s Old Age Pension and Child Grant Program 1000 Monthly transfer in Maloti 950 750 500 250 250 150 120 0 OAP 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 CGP Source: authors based on data from various budget speeches and World Bank country team. Notes: OAP refers to Old Age Pension and CGP refers to Child Grant Program. 29 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE 4 / EVOLUTION AND ADAPTATIONS Countries have adapted indexation choices based on changing contextual conditions. Drawing on 14 case studies, this section briefly traces the tra- jectories of select programs over time.26 Specifically, the section offers a framework to examine three broad choices on methods, mechanisms and frequency (figure 15). The basic framework tracks how programs moved from discretionary to automatic indexation (Argentina, Australia, and Gha- na); a case of movement in the opposite direction, i.e., from automatic to discretionary indexation (Maldives’s “conditional” indexation); and a set of hybrid models combining elements of both (Norway and Mexico). Other cases such as Uruguay show how a program can be effective while not changing its indexation approach. In terms of mechanisms, the analysis showcases practices of updating measures within a price-based index- ation framework (Belgium), as well as shifts from prices to wages (New Zealand). Furthermore, mixed approaches in Germany are discussed. And finally, two cases of shifts in frequency are presented, including a case of reduced frequency (Italy) and of augmented one (Canada). Each of these case studies offer more insights than just the outlined trends, although these experiences are by no means exhaustive. For instance, in Tajikistan benefits of the Target Social Assistance program were lastly up- dated in 2010 until the government introduced an indexation mechanism in 2020. Conversely, in Finland the non-contributory child benefit program was linked to the national pension index until March 2011, when indexation was temporary suspended from 2013-15 and later abolished in 2016 (for more info see box xx). Change within ad-hoc indexation were also doc- umented. In Libya, a case of ad-hoc indexation, the Basic Pension Ben- efit has been updated several times since its introduction in 1985: while benefits were 50% of the minimum salary in 1985, the underlying Law was amended in 2013 setting the basic pension benefit amount to 100% of the minimum wage. 26 — See Arimbi (2024a, 2024b), Aziz (2024a, 2024b), Cheng (2024a, 2024b), Nogueira (2024a, 2024b, 2024c), TMM Iyengar (2024), Trujillo (2024a, 2024b, 2024c), and Valleriani (2024). 30 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Figure 15 / Adaptation of indexation choices over time, select programs Method Mechanisms Frequency Automatic Prices More frequent Argentina AUH, Maldives Hybrid: Within New Zealand Hybrid: Canada Italy Australia AP, BP Norway prices: JS Germany Ghana LEAP housing Belgium Citizen allowance; health benefit Mexico index PBPAM Discretionary Wages Less frequent Source: authors’ original figure for this publication. FROM DISCRETIONARY TO AUTOMATIC INDEXATION Argentina The Universal Child Allowance (AUH) was launched in 2009 as Argentina’s flagship conditional cash transfer program. The scheme covers about 29% of the population and claims around a half percentage point of GDP. Ini- tially, the benefit amount per child was 180 pesos and 720 pesos for each disabled child (four times the normal amount). Subsequently, the value was adjusted on a discretionary basis via executive decree until July 2015. The Law established that AUH benefits would be updated according to the same calculation of the “mobility index” used for pensions. Such index in- cluded two components: the variation of a wage index and the variation of a tax revenue index. Between 2015 and 2017, the AUH was updated every six months. From 2017 to 2021, the estimation method of the mobility index and the update frequency were modified again to four times a year (in March, June, September, and December). This adjustment is based on a new mobility index, combining two elements: a 70% weight of annual changes in CPI, estimated with a six-month lag, and a 30% weight of annual changes in the Average Taxable Remuneration of Stable Workers (RIPTE). As observed in figure 16, the per-child benefit amount increases from ARG $180 in 2009 to ARG $4107 in 2021 (nominal terms). 31 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Figure 16 / Minimum AUH benefit amount (per child, monthly, ARG $) 5000 ARG $ 4.017 4000 3.293 3000 2.652 2000 1.816 1.412 1.103 1000 837 644 460 270 340 180 220 0 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Source: authors based on data from various government official materials (see references) Those nominal increases helped to keep up with inflation up to a certain point. For instance, inflation reached the stunning (annual percent change) rate of 49.2% in 2018 and 54.2% in 2020. The loss in purchasing power can be estimated by comparing the minimum per capita benefit as currently calculated and a hypothetical benefit indexed solely to annual variation of the general CPI. By 2021, the disparity between the minimum benefit per capita and the hypothetical inflation-adjusted benefit is estimated to have reached 26.4% (figure 17, left side). In other words, the per capita value of the benefit provided in 2021 should have been about one-quarter higher to keep up with inflation. Such gap widens to 33.9% when considering the minimum benefit amount indexed to prices in the Basic Food Basket (food prices rose at a higher rate) (right side of figure 17). 32 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Figure 17 / Comparison of AUH actual and projected CPI-adjusted benefits (left) and benefit gaps (right) 6000 ARG $ 10% 5000 0% 4000 -10% -26,4% 3000 -20% 2000 -30% 1000 -33,9% 0 -40% 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Minimum amount per capita % gap Benefit (using CPI) Minimum amount per capita (adjusted with inflation) Source: authors based on data from various government official materials (see references) Notes: benefits refer to monthly amounts. Adjusting for inflation comes at a fiscal cost. While AUH expenditures were 0.53% by 2021, switching to a general CPI-based indexation would have increased costs by 0.2 percentage points of GDP (reaching 0.74%), while using the inflation rate of the basic food basket would have resulted in a slightly higher expenditure of 0.75% of GDP in 2021. As of March 2021, the Government applied the new mobility formula voted by Congress in De- cember 2020. Unlike the previous model, the new calculation arises from the sum of 50% of the quarterly increase in the collection by the Nation- al Social Security Administration (ANSES) and 50% for salary variation of state employees (RIPTE index) of the same period. In the latter case, data from the National Institute of Statistics and Census (INDEC) or the Ministry of Labor are used, depending on which was the highest. Australia’s social pensions Between 1909 and the 1970s, indexation of benefits of Australia’s non-con- tributory pension (Age Pension program) followed a systematic approach governed by Acts and regulations. Indexation was introduced in 1932 until 1937, then reintroduced between1940 and 1944, and adopted again in 1973 for another 3 years. In 1977, the current practice of automatic indexation was eventually adopted. At the time, the price index used for reference was the general CPI. Throughout 1983-1996, four discretionary adjustments were made to these pension rates above that of the twice-yearly index- ation. Subsequently, in 1997, the practice of benchmarking Age Pension 33 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE benefits for single pensioners to the male total average weekly earnings (MTAWE) was introduced, and the partnered adult rate of pension set at 83% of the single adult rate of pension. Following the findings of the 2009 Pension Review Report, a dual indexation mechanism was adopted. This approach considers two separate price indices – the Pensioner and Bene- ficiary Living Cost Index (PBLCI) and the CPI (figure 18). Figure 18 / Evolution of Age Pension benefits 30,000 Maximum pnsion rates per annual (in Australian $) 2009 Current indexation 25,000 mechanism introduced 20,000 15,000 1909-1976 1977 Indexation of pension Twice-yearly rates based on Acts automatic indexation 10,000 and regulations to CPI introduced 5,000 0 Standard rate ($ p.a.) Partnered 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021 rate ($ p.a.) Source: authors based on data from various government official materials (see references) The PBLCI mainly serves to reflect the out-of-pocket expenses of a sub- group of pensioners, which includes higher weights for expenditure items like health and food, and lower weights for housing. Australia’s Age Pen- sion is currently automatically indexed in March and September each year, using the greater of 6-month changes in the CPI or the PBLCI. After in- dexation, the pension rate is then compared to the MTAWE, and increased further if the pension rate for couples is less than 41.76% of the MTAWE. Additionally, the eligibility criteria for the income and assets tests are also concurrently revised during indexation, preventing situations where ben- eficiaries “inflate” out of eligibility as their nominal incomes rise. Overall, estimates suggest that since the introduction of twice-yearly automat- ic indexation in 1977, the discrepancy between the real value of pension benefits (relative to 1963 levels) only varies by an average of 0.5% annually (figure 19). 34 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Figure 19 / Actual indexed benefits versus implied real value (percentage difference in actual benefits versus its implied real value) 20% Post 1976 % difference from 1963 benefits purchasing power Switch to twice-yearly automatic indexation reduced discrepancy between actual amount and 10% 0% -10% 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021 Source: authors based on data from various government official materials (see references) Ghana: a real-time shift Between its inception in 2008 and 2023, Ghana’s LEAP program only ad- justed benefits three times on a discretionary basis, namely in 2012, 2015 and 2023 (Figure 20). Starting 2024, the program is on a path toward au- tomatic indexation. At the time of writing this report, such automation was reflected in a Cabinet-level paper and it was just approved by the parlia- ment as part of the Social Protection Bill. Figure 20 / Ghana’s nominal and real LEAP benefit 90 Jan 2023 Ghana cedis Nominal benefit doubled 80 79.00 70 60 Sep 2015 50 Nominal benefit Jan 2012 increased by 25% Nominal 39.50 40 benefit trippled 31.75 30 20 10.50 10 0 2009 2009 2010 2011 2012 2013 2014 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Standard rate ($ p.a.) M02 M12 M10 M08 M06 M04 M02 M12 M10 M08 M06 M04 M12 M10 M08 M06 M04 Partnered rate ($ p.a.) Source: GoG (2023) 35 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Recent analysis shows a doubling of benefit levels would reduce the ex- treme poverty headcount by up to 0.7 percentage points.27 Assuming same coverage levels as in 2023, LEAP transfers in 2024 would cost close to an additional 0.02% of GDP, leading total spending to 0.07% of GDP. Specifi- cally, the indexation procedure would entail an annual adjustment of the LEAP benefits linked to lagged inflation to be done in October each year, expressed as: LB t+1 = LB t ( 1+ inf t ) …………… (1) Where: LB t+1 = Average LEAP Benefit per household for the next fiscal year; LB t = Average LEAP benefit per household for the current year; and inf t = 12-month annual average inflation from October of previous year to September of current year (Occt−1 to Sept) estimated from Ghana Statisti- cal Service (GSS) Consumer Price Index release. In the indexation formula (Equation 1), the use of lagged inflation series is to minimize errors and potential biases linked to forecasting inflation.28 The formula is set to reflect any additional increase in the benefits provid- ed by government prior to the indexation adjustment of the ensuing year; and to ensure sustainability of the indexation mechanism, the annual ad- justment of the LEAP benefits will be capped at 50%. Pakistan offers another example of real-time developments: while the country hasn’t’ introduced yet an automatic indexation mechanism, it is in the process of exploring options. Simulations indicate that anchoring benefits to CPI and other methods would not entail significant fiscal impli- cations (box 1). 27 — Tesliuc, C., Corral, P., Gupta, S. and Ampredu, C. (2024) “Ghana Livelihood Empowerment Against Poverty (LEAP): Assessment of doubling LEAP benefit level in 2024”. Mimeo. Accra. See also Government of Ghana (2023, 2024). 28 — GoG (Government of Ghana) (2023) “Indexation Mechanism of Benefits Under the Livelihood Empowerment Against Poverty (LEAP) Cash Transfer Programme”. Ministry of Finance. Accra. 36 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Box 1 / Indexation in Pakistan The unconditional cash transfer (UCT), Kafaalat, under the Benazir Income Support Pro- gram (BISP) was launched in wake of the global financial crisis in 2008 to provide con- sumption support to poor and vulnerable families to mitigate the impact of soaring food and fuel prices. Since its launch, BISP increased the amount of UCT benefits several times, especially during 2023-2024, but its real value remains lower than what it was in 2008 at the time of its launch: between 2008 and 2023, the nominal value increased by 250 per- cent, but the real value decreased by 16 percent. So far, whenever Kafaalat benefit amount was increased, the decision was rather ad hoc and was based on the available fiscal space, tradeoff between coverage and adequacy, and other political economy issues. However, in 2022 the Ministry of Finance and Revenue constituted a committee to regularly review cash transfer benefit levels. Simulations suggest that “… had an indexation policy (…) been introduced at the time of BISP’s launch, the total spending will still have remained below 0.6 percent of the GDP. This shows that the fiscal burden of such a policy will not become too high to make it unaffordable.” 0,50% Cost 0,46% 0,43% 0,45% 0,39% 0,40% 0,35% 0,30% 0,25% 0,25% 0,20% 0,15% 0,10% 0,05% 0,00% FY2009 FY2010 FY2011 FY2012 FY2013 FY2014 FY2015 FY2016 FY2017 FY2018 FY2019 FY2020 FY2021 FY2022 FY2023 FY2024 FY2025 FY2026 FY2027 FY2028 UCT spending (%GDP) Scenario spending (%GDP) Source: Majoka (2024). Data on indexation refers to the highest cost between updating benefit level based on CPI or updating the transfer based on expenditure data from household surveys (i.e., an increase in generosity by 1 percentage point/year). 37 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE CHANGE IN METHOD: FROM AUTOMATIC TO DISCRETIONARY INDEXATION Maldives: from “conditional automation” to discretionary indexing The old age Basic Pension (BP) program in the Maldives includes non-con- tributory transfers to 82% of seniors and claims about 1.2% of GDP. As per the original 2009 Pension Act, BP is set at MVR2,000 per month. The Act requires a ministerial committee to review the benefit amount annual- ly, mandating that the benefit be uprated if the cost of living increased by more than an annual rate of 5% (as indicated by annual national CPI). Figure 21 traces the recent evolution of indexation in the country. In 2012, the benefit amount was uprated to MVR2,300 in line with high inflation in 2011 exceeding the 5% threshold. However, the BP benefit amount was not uprated again in the six years between 2013 and 2019, amid relatively low inflation over the period that did not breach the 5% threshold. Nonetheless, in 2014, as part of a presidential campaign pledge, a separate program called Senior Citizen’s Allowance (SCA) was introduced, which effectively acts as a “top-up” benefit to the existing MVR2,300 BP benefit – bringing the combined BP and SCA benefit amount to MVR5,000. Figure 21 / Evolution of BP benefits in the Maldives 6000 2014: 2019: BP Monthly Benefit Amount (in Maldivian Rufiyaa) A non-BP SCA “top-up” SCA benefits were consolidated into BP, allowance introduced increasing BP benefits to MVR5,000 4000 2012: BP monthly 2010: benefit increased BP initial amount to MVR2,300 set a MVR2,000 2000 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Source: authors based on data from various government official materials (see references) In 2019, the Pension Act of 2009 was amended to consolidate the SCA “top- up” allowance directly into BP, thereby discretionarily raising the guaran- teed BP benefit to MVR5,000. De facto, eligible pensioners have received MVR5,000 in monthly benefits (first SCA and BP combined, and later as BP) since 2014, despite inflation not exceeding 5% over the period. Furthermore, the review process was changed to a three-year period. Notably, the 2009 provision mandating the uprating of BP benefits “if the cost-of-living indica- 38 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE tor exceeded 5%: was removed, indicating a transition from the “conditional indexation” in the original Act to one where the benefit-setting is at the dis- cretion of the inter-ministerial committee or the President. In terms of purchasing power over time, the real value of Basic Pension (BP) benefits – excluding the SCA “top-up” in 2014 – experienced a gradu- al erosion due to inflation from 2011 to 2018, particularly because BP ben- efits were unchanged in nominal terms from 2013 to 2018 (Figure 22). Con- sequently, by 2018, the purchasing power of the BP benefit had decreased by approximately 20% compared to when it was initially set in 2010. How- ever, the consolidation of SCA benefits into BP in 2019 significantly in- creased monthly BP benefits, and as of 2023, current BP benefits are about 67% above original 2010 benefits in real terms. If instead, the BP benefit amounts were re-indexed starting from 2014 (with SCA inclusion) and 2019 (BP only), the real value by 2023 shows a decline of approximately 4.1% and 9.5% respectively. Figure 22 / Percentage real shortfall from original BP benefit amount real terms Percentage shortfall from original benefit amount in 75% 50% 5% 0% BP+SCA: indexed to 2014 levels, -9.5% BP only: indexed -25% to 2019 levels, -4.1% BP only: indexed 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 to 2010 levels, +67% Source: authors based on data from various government official materials (see references) NO CORRECTION Uruguay’s steady course The Family Allowance of the Equity Plan (AFAM PE) in Uruguay is a con- ditional cash transfer program created in January 2008. It reaches about 45% of the population cohorts aged 0-17 at a cost of 0.48% of GDP. The benefit amounts are updated annually based on CPI variation. This adjust- ment mechanism has not changed since the program’s inception. For most of the program’s life, AFAM PE’s CPI-anchored benefits have performed in 39 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE line with alternative price indices, e.g., the average annual inflation rate, annual price fluctuations in food and non-alcoholic beverages, and vari- ations in the cost of the essential goods (rural and urban) (Figure 23). Yet the pandemic year of 2020 marked a discontinuity, with rising food prices exceeding the increase in overall CPI and forming a (relatively minor) gap of about 5% (figure 24). Figure 23 / Comparison of inflation-adjusted benefits using multiple indexes vs actual AFAM PE payouts (monthly amount in UYU $) 2500 AFAM PE montly benefit amount (UYU $) UYU 2.066 UYU 2.063 2000 UYU 2.065 UYU 1.923 UYU 1.922 1500 1000 500 Actual 0 Indexed - Prices of basic food basket (urban) Indexed - Annual average inflation 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Indexed - Prices of basic food basket (rural) Source: authors based on data from various government official materials (see references) Notes: AFAM PE stands for Family Allowance of the Equity Plan. Figure 24 / Benefit gaps: inflation-adjusted benefits using multiple indexes vs actual AFAM PE payouts AFAM PE benefit gap 4% 0% -4% -8% 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Actual average inflation Inflation of food and non-alcoholic beverages Prices of basic food basket (rural) Prices of basic food basket (urban) Source: authors based on data from various government official materials (see references) Notes: AFAM PE stands for Family Allowance of the Equity Plan. 40 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE HYBRID Norway: indexation plus discretionary top-up Norway’s housing benefit program (bostøtte) is a non-contributory benefit that is given to household with low incomes and high housing expenses. As of December of 2023, more than 152,000 households, roughly 5.8% of Nor- way’s population, received the allowance. The housing allowance is calculat- ed as 73.7% of the difference between housing expenses and a households’ “contribution.” The calculation is based on the following formula: Housing allowance = 0.737 * (approved housing expenses – household contribution) The “approved” housing expenses have upper ceilings, which vary by mu- nicipality. The household “contribution” depends on income and family size. The higher the income is, the larger the contribution.29 The benefit amount is indexed automatically on June 1st of each year based on the av- erage CPI of the preceding twelve months. The price adjustment mecha- nism according to CPI was introduced in 2017. The housing allowance may have not kept the pace with the growing cost of living. Figure 25 illustrates that annual average housing allowance compared with average month- ly rental prices of a two-bedroom house across Norway: the latter have nearly doubled in a decade, rising from approximately NOK 4,000 in 2012 to more than NOK 7,000 in 2021. From January to May 2023, recipients were granted a discretionary top up benefit to mitigate higher electricity bills (NOK 1,500 per month), while an extra NOK 1,000 was approved for the same months of 2024. Figure 25 / Average benefits and national average rental price (2-bedroom housing) Average annual houseing allowance vs. Average monthly rental prices (two-bed) 10000 9800 7500 6200 5000 2863 2500 2132 0 Average Benefit Amount 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Average Monthly Rent for 2 Rooms in Norway Source: authors based on data from various government official materials (see references) 29 — See https://rm.coe.int/nor-ad-hoc-report-on-cost-of-living-crisis-2023/1680ae60b8. An online eligibility calculator can be found here: https://husbanken.no/english/housing-allowance/am-i-entitled/ 41 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Mexico’s “unspecified automatic” model Mexico’s experience was characterized by a particular form of automatic mechanism, that is, there is a clear legislated provision indicating the need for annual updates in transfer size, but with no indication on how such re- visions should be undertaken.30 This particular case relates to the Pension for the Well-being of Older People (PBPAM) social pensions program es- tablished in 2019. The program was preceded by a range of interventions involving shifts in eligibility and transfer modalities (box 2). Box 2 / The road towards PBPAM social pensions In 2003, the “Program for the Care of Older Adults” focused on individuals aged 60 and older and was means-tested. In 2004, it transitioned to a “food assistance” program until 2006. By 2007, the social pension was re-established with the “Program for the Care of Older Adults Aged 70 and Older in Rural Areas” (Program 70 and More). The scheme was limited to communities with up to 2,500 inhabitants and targeted older adults in poverty, vulnerability, and marginalization. Initially covering slightly over one million people (16.2% of those aged 65 or older), in 2009 the program extended its reach to communities to up to 30.000 inhabitants, thereby expanding coverage to over three million individuals (41% of those aged 65 or older) in 2012. In 2013, the program was renamed “Pension Program for Older Adults” (PAM), lowering the eligibility age to 65 and targeting older adults without income from contributory retirement or pension plans. This significantly increased cov- erage to 4.8 million older adults (62.6% of the elderly population). The program retained its name and parameters until 2018, when it provided nationwide coverage and reaching 5.1 million older adults (53.8%). In 2019, the PAM was replaced by the “Pension for the Well-being of Older People” (PBPAM) and expanded its target population to include indig- enous individuals aged 65 and older, those aged 68 and older, and individuals aged 65 to 67 incorporated in the Active Beneficiary Registry of the program. Notably, this program eliminated eligibility criteria based on poverty or income leading to an increase in program coverage by nearly 3 million: in 2019, PBPAM provided social pensions to over 8 million people, accounting for 89.9% of the elderly. In 2021, the program was extended to inclsude the Afro-Mexican population aged 65 and older. Additionally, on July 7 of that year, an Amendment Agreement was issued, unifying the target population to individuals aged 65 and older and making this benefit universal for individuals in this age range. Consequent- ly, the program’s coverage increased to 9.6 million in 2021, reaching 93.5% of the target population. In 2022, PBPAM covered almost the entire target population, encompassing 98.9% of older adults in Mexico. 30 —The operation manual mentions the requirement for annual benefit adjustments, but changes are determined by the executive branch. 42 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Until the introduction of PBPAM, adjustments to benefit amounts of its pre- decessors were discretionary or due to changed configuration to the pro- grams (figure 26). In 2019, the benefit amount increased to approach the extreme poverty line. In 2020 and 2021, it rose in line with the inflation rate, and in 2022 and 2023, significant increases were implemented to bring the social pension closer to the poverty line. Thus, since 2019 changes have not adhered to a predetermined pattern. Figure 26 / Social pensions benefit evolution 3000 (MXN $) Monthly minimum amount per older adult 2500 2000 1500 1000 500 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2007-2012: 2012-2018: 2018-Extreme 2019-present: 2022 - Poverty line Felipe Calderón Hinojosa's Government Enrique Peña Nieto's Government poverty line Rural: Andrés Manuel López Rural: MXN 3,051.80 Program for the Care of Older Adults Pension Program for Older Adults MXN 1,208.47 Obrador's Government Urban: MXN 4,246.06 Aged 70 and Older in Rural Areas (PAM) Urban: MXN 1,586.96 Pension for the Well-being (Program 70 and More) of Older People (PBPAM) Source: authors based on data from various government official materials (see references) When comparing a hypothetical CPI-indexed value of social pensions and the actual benefit value, there is no gap in 2020 and 2021 (figure 27). Those were in fact the years when, as just mentioned, the actual value of the so- cial pension was adjusted using the inflation rate. In 2022, a year when the pension was discretionarily increased by MZN 575, the actual value of the pension surpassed the inflation-indexed value. In 2022, the actual value would be 32.8% higher than the indexed value, and in 2023, it would be 53.6% above it. Those adjustments even exceeded values adapted to food inflation. 43 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Figure 27 / Difference between actual and inflation-adjusted social pension benefits (% difference) 53,6% 39,5% 60% Benefit gap (difference b/w actual vs. inflation adjusted social pension) 32,8% 24,7% 40% 20% 0,0% 0,0% 0,1% 0% -2,1% -3,6% -3,6% -9,5% -12,8% -20% -12,7% -16,2% -16,3% -19,2% -13,2% -22,3% -19,4% -23,7% -18,3% -26,2% -16,6% -26,1% -18,3% -27,5% -21,0% -30,7% -40% -26,0% -36,1% -60% Inflation 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Basic food basket Source: authors’ estimation using data from from Diario Oficial de la Federación (DOF), Legal acts that create the social pension for the years 2007, 2013a, 2013b, 2014a, 2014b, 2015, 2016, 2017, 2019a, 2019b, 2020a, 2020b, 2021a, 2021b, 2022, CONEVAL (2023) and INEGI (2023). Note: the 2019-2023 is estimated using the re-indexed values. The Basic food basket is for rural areas. Expenditure projections indicate that if annual automatic indexation had been introduced throughout 2007-18, program costs would have increased from 0.15% of GDP to 0.18% (and to 0.22% when considering inflation an- chored on a basic food basket) (figure 28). For the period 2019-21, with re-indexed values and updates to the benefit amount based on inflation rates, projected and actual expenditures align with actual expenditure. It is likely that savings could also have been incurred in 2022-23 if automatic indexation to CPI has been established. 44 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Figure 28 / Cost of actual and inflation-adjusted programs 0,60% Expenditure (% of GDP) 0,56% 0,55% 0,55% 0,50% 0,40% 0,30% 0,22% 0,20% 0,18% 0,15% 0,10% 0,00% Actual Inflation 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Basic food basket Source: CEPAL Non-Contributory Social Protection Programs Database and authors’ estimates using INEGI Data on GDP. Notes: the 2019-2023 expenditure is estimated using the re-indexed values. Expenditure excludes administrative costs. The Basic food basket is for rural areas. CHANGE IN FREQUENCY (WITHIN AUTOMATIC INDEXATION) Italy: from more to less frequent indexation The Italian non-contributory Incapacity Pension, launched in 1971, cu- rrently covers about a million people, or roughly 2.7% of the population. The indexation framework for core social protection programs31, including the Incapacity Pension, moved from annual (1969), to quarterly (1984), to biannual (1986) to annual (1994) uprating frequencies – with several other changes occurring over such timeframe (Figure 29). Figure 29 / Evolution of automatic indexation Automatic Indexation 30 April 1969: 1975: 1984: 1986: 1994: 1996: 2013: 2024: annual based on wage of blue-collar quarterly based biannual based annual based annual based amount varies switch from “sliding-wage scale” workers on “sliding-wage on “sliding-wage on CPI (FOI) on CPI (FOI) of by class of four to six class (Law N. 153) (Law N. 160) scale” scale” (Amato reform previous year pension income income catego- (Law N. 730) (Law N. 730) N. 503/1992) (Law N. 724/1992) (Law N. 147) ries (Budget law 2023) Basic Program Information March 1971: 1980: civil disability accompaniment pension allowance established established (Law N. 118) (Law N. 18) 1969 1970 1975 1984 1986 1994 1996 2013 2024 Source: authors based on data from various government official materials (see references) 31 — In Italy, there are a number of relevant programs pertaining to different areas of social support that could be explored and scrutinized by further research. A few examples are the Single and Universal Allowance for Children (Assegno Unico e Universale per i Figli - AUU), Inclusion Allowance (Assegno d’Inclusione - ADI), Support for Training and Work (Supporto per la Formazione e il Lavoro - SFL), and the Social Allowance (Assegno Sociale). 45 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Automatic indexation of pensions was firstly introduced in Italy in 1969, with benefit amounts increased on January 1st in line with the cost of living or “sliding-wage scale.” In 1975, it was established that pensions would in- crease based on the percentage change of the minimum wage of blue-co- llar workers. However, this mechanism was short lived. Starting from 1984, indexation became quarterly (1st of February, May, August and November) and it was based again on the cost of living for “sliding-wage scale.” Notably, the law introduced for the first time ever an indexation mechanism which varied according to the cumulative amount of pension received; a 100% in- dexation would apply for pension amounts that were double the INPS mini- mum amount, 90% for amounts between two and three times the minimum and 75% indexation for amounts over three times. While in 1986 indexation turned biannual (1st of May and November), it is in 1994, following the Amato reform of 1992), that indexation was made annual again (1st of November) and it switched from a “sliding-wage scale” to the index of consumer pri- ces of blue-collar families, also called FOI index, as determined by National Institute of Statistics (ISTAT). Finally, in 1994 it was established that start- ing from 1996, indexation would be carried out once a year (1st of January) based on the FOI index for the previous year. In 2013, indexation was reformed and based on bands or classes of amounts, that is, a pension is revalued with only the rate corresponding to the bracket or class in which the cumulative pension amount falls. The different income classes are calculated based on the minimum INPS set by law. As a result, the bands-system produces diverse pension growths across categories through time, which can translate into a loss in real terms that is (i) higher the smaller is the percentage of indexation com- pared to a 100% adjustment, (ii) the more the years of pensions, and (iii) worse the higher is inflation. Recent reforms are presented in box 3. Box 3 / Recent indexation reforms in Italy Starting in 2022, indexation applied 100% for cumulative amounts up to four times the minimum INPS; 90% to amounts above four and up to five times the minimum INPS and 75% to amounts over five times the minimum. Recently, with the budget law approved in November 2023 the system moves from four to six bands: • 100% indexation up to 4 times minimum INPS • 85% indexation above 4 and up to 5 times minimum INPS • 53% indexation above 5 and up to 6 times minimum INPS • 47% indexation above 6 and up to 8 times minimum INPS 46 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE • 37% indexation above 8 and up to 10 times minimum INPS • 22% indexation above 10 times minimum INPS Automatic indexation follows a two-phase methodology. First, an adjustment takes place in January, based on the provisional FOI index of the previous year (i.e., at time t-1), which is subject to subsequent adjustment in January of the following year based on the final FOI index (at time t). The end-of-year adjustment is calculated comparing the provisional FOI index with the final inflation rate. The Ministry of the Economy and Finance, jointly with the Ministry of Labor and Social Policy, are in charge of determining the revaluation by November 20th of the current year to be applied the following January. The adjustment can thus be: • positive, with an additional amount to be paid on the pension, if the final equalization is higher than the projected equalization; • negative, with an amount to be deducted from the pension, if the final equalization is lower than the projected one; • nil, if the final equalization is the same as the forecast equalization: in the latter case, there is no change in the pension check. Indexation was key for preserving purchasing power. Analysis conducted for this report shows that if the 2012 value had not been adjusted to infla- tion, the erosion of the transfer would have resulted in a 45% loss in pur- chasing power by 2023. Besides the FOI index, ISTAT also produces two other indices, which can be used here for comparing FOI performance: NIC and IPCA. NIC and FOI are based on the same basket of goods (i.e., 1,885 representative goods), although each good is given a different weight based on their relevance in terms of consumption for the underlying pop- ulation considered. For NIC, the population reference is the entire popu- lation on the Italian territory, while FOI only refers to the consumption of families headed by an employee (from the non-agricultural sector). On the other hand, the IPCA index was developed to assure an inflation measure comparable at the European level. FOI and NIC-adjusted benefits follow the same pattern with no major discrepancies (NIC benefits being 1.3% higher), while between 2021 and 2023 IPCA-calibrated benefits were between 5% and 13.4% higher than FOI’s (Figure 30). 47 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Figure 30 / FOI (actual) versus hypothetical NIC and IPCA-adjusted benefits 1,5% 1,7% Benefit gap (actual benefit amount vs.NIC and IPCA-adjusted benefits) 2% 0,8% 0,4% 0,4% -0,2% -0,1% -0,3% -0,5% -0,7% -0,8% -0,9% -1,0% -1,1% -1,1% -1,2% -1,3% -1,2% -0,8% -1,3% -0,3% -1,4% -1,5% -0,4% -2% -1,1% -2,2% -1,1% -2,4% -1,2% -4,1% -1,3% -5,0% -6% -10% -1,3% -9,2% -14% -1,3% -13,4% FOI-NIC 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 FOI-IPCA Source: authors based on Istat and Inps data Canada: from less to more frequent indexation In Canada, the non-contributory social pension, i.e., the Old Age Security (OAS) program, has its roots in a 1927 means-tested pensions program. In 1967, the program eliminated its means-test and became solely age- based. OAS currently covers almost the entire (96-98%) elderly population (of at least 65 years of age).32 The program was indexed annually until 1973, when it changed to quarterly indexation due to the high level of inflation. As such, OAS benefits are reviewed each year in January, April, July and October to ensure they reflect cost of living increases, as measured by the CPI. Monthly payment rates will not decrease if the cost of living declines. For example, to calculate the rate of increase on January 1st, 2024 (which define benefits from January to March 2024), the difference between the average monthly CPI from August to October 2023 and the average CPI from May to July 2023 is divided by the latter average. The result is round- ed to three decimal places and then expressed as a percentage. Indeed, on January 1st, 2024, OAS cash transfers for the reference benefit period (column (a) on table 1) were increased by 0.1% since the average CPI was 123.1 from August to October 2023 (column (b) on table 1), while it was 123.0 from May to July 2023 (column (c) on table 1). As such, the formula is a = (b – c)/c = 0.001, or 0.1%. 32 — About one-third of OAS beneficiaries received an income-tested program, the Guaranteed Income Supplement. 48 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE Table 1 / Benefit, comparison, and initial periods for OAS indexing Benefit Period (a) Final Comparison Period (b) Initial Comparison Period (c) January to March 2024 August to October 2023 May to July 2023 April to June 2024 April to June 2024 August to October 2023 July to September 2024 February to April 2024 November 2023 to January 2024 October to December 2024 May to July 2024 February to April 2024 Preliminary analysis shows that quarterly indexation may be more effec- tive in protecting against inflation than other frequencies. For instance, an- nual indexation may close the gap of benefit loss at the end and beginning of the year, but its performance deteriorates throughout the year. Similarly, monthly indexation (using the current and previous month’s CPI) may not reflect the seasonal element of the last quarter of the year when compared to quarterly or annually indexed benefits (figure 31). Figure 31 / Benefit loss due to inflation across different indexation frequencies 10% and IPCA-adjusted benefits) Benefit gap (actual benefit amount vs.NIC -10% -30% -50% -70% -90% Benefit Loss (CP I Monthly) Benefit Loss May May Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar Jul Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar Jul Sep (CP I Annually) Benefit Loss 2013 2014 2015 2016 2017 2018 (Current Indexation) Source: authors based on data from various government official materials (see references) CHANGE IN MECHANISM: FROM PRICES TO WAGES New Zealand The Jobseeker Support Allowance is non-contributory unemployment as- sistance program introduced in 2013 as part of a wider set of programmat- ic mergers within the country’s social protection reform. It currently reach- es about 5.9% of the working age population at a cost of around 0.9% of GDP. The benefit amounts have increased on yearly basis, but the largest 49 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE adjustments have taken place in 2016 (changes in CPI)33 and, especially, starting from 2020 when the program switched indexation from prices to wages (figure 32 and 33).34 Figure 32 / Evolution of Jobseeker Support benefit amount 750 Axis title 600 450 300 150 JS 25+ JS Couple (w/o children, total) 0 JS Couple (w/o children, each) JS Couple (w/ children, total) 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 JS Couple (w/ children, each) Source: authors based on data from New Zealand’s Work and Income Notes: the benefit amount represent Net weekly rate (i.e., after tax at “M”) as of April 1 each Figure 33 / Evolution of automatic and ad hoc adjustments Other ad hoc adjustments One-off Inequality and child Welfare Covid-19: poverty (Budget assistance $25 2021): 2 stage adjustment package: 0.98% Indexation Since 1992: Starting 2016: Starting 2020: CPI (all groups) Changed to CPI (all groups, Changed to Net Average exc. cigarettes and other wages (full time equivalent) tobacco products) Basic Information July 2013: Jobseeker Support started 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Source: authors based on data from various government official materials (see references) 33— The Jobseeker program existed before 2013 under different names and forms. Since 1991, it was anchored on CPI (all groups), before using a narrower CPI in 2016 (all groups, excluding cigarettes and other tobacco products). 34 — There are other ad hoc adjustments that have taken place; all such adjustments are made through a Cabinet decision or convention. Such ad hoc adjustments happen, in addition to (i.e., on top of) regular annual automatic adjustments. As highlighted in figure x, over the past decade there have been several instances where benefit amounts experienced such top-ups: (i) during Covid-19; (ii) to reduce inequality and child poverty (2021, 2022); and (iii) in 2023 in response to cost- of-living crisis. 50 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE The price-to-wages shift was primarily driven by fact that, for almost three decades, the net average wages have grown faster than inflation.35 Specifi- cally, the regular annual increases in Jobseeker Support’s benefit amounts happen on April 1 through an Annual General Adjustment statutory pro- cess. Starting in 2020, the benefit amounts are adjusted annually based on the percentage upward movement of average wages (measured using “net average ordinary time weekly earnings”). In particular, the benefit amounts are adjusted each year on April 1 based on the year-on-year percentage increase in net average wages. Such percentage increase is computed by comparing net average wages in the last quarter of the previous year (Oc- tober-December) to the same period (i.e., October-December) of the year immediately preceding it. For example, the benefit amounts for “Year t” ( ) will be adjusted based on the percentage increase in net average for the last quarter of “Year t-1” () when compared to the last quarter of “Year t-2” (). wages If the percentage increase for this period was 5%, then the benefit amount for Jobseeker Support Program will be increased by 5% for . Recent analysis on the Jobseeker program (and other schemes) shows that from 2020, the anchoring on wages, combined with ad-hoc increases, has reversed long-term declines in benefit adequacy as a share of wages36 (figure 34). Figure 34 / Adequacy of Jobseeker Support as percentage of net average wages 100 DPB: Domestic Purposes Benefit SPS: Sole Parent Support 80 UB: Unemployment Benefit JSS: Job Seeker Support IB: Invalids Benefit SLP: Supported Living Payment 60 WEP: Winter Energy Payment BS: Best Start AS: Accommodation Supplement 40 20 NZ S Couple DPB/SPS, 2 ch DPB/SPS, 1 ch NZS SLA 0 UB/JSS couple IB/SLP single 1945 50 55 60 65 70 75 80 85 90 95 00 05 10 15 20 2025 UB/JSS single Source: Perry (2022) 35 —The change was recommended by a Welfare Expert Advisory Group. 36 — See Perry, B. (2022) “Child Poverty in New Zealand”. Ministry of Social Development. Wellington. 51 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE MIXING WAGES AND PRICES Germany’s supplementary indexation As part of reforming its iconic Hartz IV system, in 2022 Germany intro- duced the Citizen Benefit Program. Geared to support unemployed and low-income households, the scheme reached about 6.5% of the population in 2023. Its introduction was underpinned by a new formula for benefit in- dexation which ramped-up benefits significantly: between 2012 and 2021, annual benefits had increased by an average of about 1.9%, while in 2023 it rose by 11.8% (figure 35). Figure 35 / Standard Hartz IV benefits and annual CPI 12 % change in CPI and Benefit amount 9 6 3 0 Annual Change in CP I 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 % Increase in Benefit Amount Source: authors based on data from various government official materials (see references) The Hartz IV program used an automatic adjustment mechanism estab- lished by law. This involved a “basic update” where standard benefits are updated to inflation by a mixed index: 70% of such index rate was based on the average prices, while the remaining 30% was anchored on trends in wag- es and salaries. The Citizen Benefit Program maintained the basic update and compounded it with a supplemental component: such step adjusts the standard amount based on the most recent or current inflation data.37 CHANGE WITHIN PRICES Belgium’s health index In the case of Belgium, the Guaranteed Minimum Income Benefit for the El- derly (GRAPA) program moved in 1994 from the historical Retail Price Index, firstly introduced in 1971, to the Smoothed Health Index. The Smoothed index is the average of the Health Index over the past 4 months, and then multiplying it by a factor of 0.98. The Health Index is used for the index- 37 — See https://www.bmas.de/SharedDocs/Downloads/EN/PDF-Publikationen/a430e-buergergeld-englisch-pdf. pdf?__blob=publicationFile&v=5 52 04 — EVOLUTION AND ADAPTATIONS KEEP THE PACE ation of housing rents, and it is calculated by removing from the CPI highly volatile (e.g., gas and gasoline) or unhealthy products (e.g., cigarettes and alcohol). Indexation only occurs when the index reaches a predetermined threshold index (also known as central index). The threshold’s value in- creases by 2% per year (e.g., the threshold index for 2023 was 125.6). 53 05 — STYLIZED FRAMEWORK KEEP THE PACE 5 / STYLIZED FRAMEWORK The previous sections of this report have laid out a set of analytical con- siderations for indexation (sections 1 and 2) and a wide range of practi- cal experiences (sections 3 and 4). But how can these observations inform decision-making process? This section offers a basic framework bringing together key issues examined in the report with a view of distilling core sa- lient factors for considering indexation. Since indexation is about changing or adapting transfer size to evolving circumstances (“vertical expansion”, left hand of figure 36), then the es- tablished framework of adaptive social protection (ASP) could present a relatively familiar set of dimensions for pondering indexation choices. These include a set of quandaries around data, programs, institutions, and financing (right hand of figure 36). We hereafter offer a discussion on how select elements of those four thematic buckets may help navigate whether indexation should be established automatically or discretionarily. Figure 36 / Indexation as adaptive social protection Benefit amount Vertical Data and Programs expansion information INDEXATION Regular Regular social Horizontal Finance Institutional benefit(s) protection expansion arrangements amount system and parameters partnerships Population Source: adapted from Bodewig et al (2021) 54 05 — STYLIZED FRAMEWORK KEEP THE PACE Starting from the financing dimension, policymakers would need to care- fully consider the “cost of action”: this involves clear fiscal implications (like those shown for Argentina), albeit the cost increase for benefit adjust- ments would hinge on program design. As discussed in the case of Gha- na, a doubling of benefits would not result in a doubling of program costs, since benefits only absorb part of program expenses (in that case, costs would rise by 40%). In the case of Pakistan, preliminary simulations report- ed in box 1 indicate that additional costs would be affordable. Conversely, fiscal costs from automatic indexation should be weighed against the “cost of inaction”. For instance, evidence shows that cash transfers can generate sizable impact; as such, the lack of augmented cash transfers can be interpreted as a negative impact or forgone impact. If for every dollar of cash transfer an additional $1.4 are generated in the economy,38 then it is plausible to assume that for every $1 in missed in- dexation would come at a potential cost of forgone $1.4 in economic mul- tipliers. Recent studies have estimated those prospective income losses.39 Furthermore, in addition to losses in human capital mentioned in section 1, possible missed cash adjustments that let purchasing power erode (“cost of living crisis”) could lead to instability: where social discontent preexists, high food prices can spark social unrest, like it recently occurred in Kenya and the earlier Arab Spring.40 The institutional dimension of indexation is crucial: this aspect under- scores how political aspects can play a key role in establishing automatic stabilizers. A discretionary system is more flexible and visible, including providing opportunities for “announcements” and space for claiming credit by political actors. Adjustments can be strategically timed to coincide with elections, thereby playing a potentially important role in electoral politics. Some categories of recipients, like seniors participating in social pension programs, can exert their political power for demanding transfer size in- creases, hence offering an explanation for why social pensions are the most indexed form of cash transfers. An automatic system of indexation would essentially curb those political interferences significantly. This is not to say that any discretionary system 38 — Gassmann et al (2023) 39 — In Uganda, Kagin et al (2024) quantified the negative multipliers stemming from halving transfers to refugees: such cut would affect both refugees and host communities, with income losses among the former ranging between 40-49%, while those for host households are about 22% (ibid, p.5). 40 — Barrett (2013), Lagi et al (2011). For Kenya, see https://www.cnn.com/2023/07/20/africa/kenya-cost-of-living-pro- tests-explainer-intl/index.html 55 05 — STYLIZED FRAMEWORK KEEP THE PACE caters for clientelist practices – also discretionary practices can base their decisions on data as shown by the many committees that meet yearly to decide on transfer size. Yet, automatically adjusted transfers introduce a fundamentally different approach based on a transparent “contract” with citizens and anchored on clear, consistent technocratic criteria. Importantly, discretionary top-ups can complement automatic provisions, hence maintaining a potential element of political signaling even when adjustments are automated. In this regard, gains can still be tangible for those introducing automated indexing, with technocratic upgrades that could even spark political gains in the longer run (politicians can claim credit for the “system” as opposed to “idiosyncratic transfers”). In terms of data, benefits from automatic indexation could be considered a form of anticipatory action: by knowing how much people are expect- ed to be paid, transfers become more reliable. Beneficiaries can plan upon expected transfers, which could help preserve assets and foster entre- preneurial risk-taking. As other forms of anticipatory, insurance-oriented measures early action in the form of automatic indexation presents limita- tions: among them is basis risk, or the possible limited correlation between changing circumstances (in this case changing prices) and needs. In other words, the way in which centrally-collected data, suboptimal indicators or composite indexes may not accurately reflect inflation “on the ground”. Finally, from a program standpoint, there are sizable benefits in automatic indexation. Coherence is one of them: this includes system-wide synchro- nization with social insurance and labor market institutions (where these are relevant), since they tend to more typically display indexation features. This would help make the social protection system as a whole more coher- ent an integrated. Technical drawbacks of integration can be equally compelling: technical complexity of calibration, maintenance, and revisions entailed by automat- ic indexation should not be downplayed. As section 4 has shown, the way in which indicators are chosen and used require a core set of administra- tive capabilities. Under what circumstances should countries consider indexation, in what way and under what conditions? Figure 37 lays out a stylized framework drawing a set of suggestions on the appropriateness of automatic index- ation. The basic parameters include the level of inflation in a country and 56 05 — STYLIZED FRAMEWORK KEEP THE PACE the degree of “maturity” of its adaptive social protection system (as for ex- ample proxied by the social protection “stress test” score encompassing the four ASP dimensions displayed in figure 36).41 Figure 37 / Illustrative framework for considering automatic indexation Where should High automatic indexation be considered? Automatic indexing Consider some level as first-best option, of indexation (e.g. integrated with other within certain limits), indexed components ponder likely high cost (e.g. pensions) at of both action and low/manageable risk inaction MATURITY OF ASP SYSTEMS (e.g. stress test score) Plausible to consider Consider indexation, some basic version but gauge the risk of of automatic indexation overheating strained fiscal and delivery capabilities (benefits from indexing may be short-lived) Low Low LEVEL OF INFLATION High Source: authors. The framework suggests that in contexts with low inflation and relatively low maturity (low stress test score), countries could cautiously introduce an automatic indexation, possibly based on clear and simple rules. Gha- na, Lesotho, and to some extent Pakistan, are some possible examples. In similar contexts of low inflation, but characterized by relatively higher de- gree of ASP maturity, automatic indexing should probably be ingrained in countries’ systems. This might be the case of several high-income coun- tries reviewed in the report’s section 4. As inflation ramps up, trade-offs in indexation become harder. In the case of countries on the lower end of the stress test, for example, an automatic indexation could still be con- sider; yet, its fiscal implications could exert considerable pressure, while benefits upgrades could be short-lived and rapidly wiped out by inflation. A number of countries facing dire macroeconomic conditions combined with high needs may fit this quadrant (e.g., Lebanon, Venezuela). Finally, the up- per right quadrant involves countries with a relatively sophisticated social protection system and concomitantly undergoing complex macro crises. Some countries in Latin America, like Argentina and to a lesser extent Bra- zil, may belong to this category. In those contexts, automatic indexation could be considered, but perhaps with novel variants that introduce some “safety valves” that could align their introduction to the state of fiscal con- 41— Bodewig et al (2021). The stress test scores countries’ ASP with values from 0 to 5 (including decimals). Over 40 countries have applied the stress test, with a variant being under development for energy subsidy reforms. 57 05 — STYLIZED FRAMEWORK KEEP THE PACE ditions – or the adoption of some blend indexation option that combined automatic and discretionary indexation based on scenarios. This may open up a novel operational research agenda for ensuring that the stark tradeoffs in such contexts – i.e., the cost of action and inaction are high – are properly pondered and reflected in indexation mechanisms that can evolve as conditions improve (or deteriorate). 58 06 — CONCLUSIONS KEEP THE PACE 6 / CONCLUSIONS The indexation of benefits represents a key and underexplored dimension of the adaptive social protection (ASP) agenda. While considerable atten- tion has been paid to coverage expansion as a core function of ASP, this report argues that indexation can be fruitfully framed as a novel feature of making social protection systems more adaptive. Through indexation, the adequacy of cash transfers can evolve – or “keep the pace” – with chang- ing conditions. This report applies an ASP framework to support policy- makers in navigating trade-offs in indexation, including presenting new data and experiences to inform whether and how indexation could be cali- brated in different contexts. Indexation practices are more prevalent and dynamic than often assumed. This report offers a novel stocktaking comprising of 232 non-contributory cash transfer programs across 158 countries. These programs, which en- compass unconditional cash transfers, conditional cash transfers, public works, and social pensions, are tracked using 16 indicators for a total of 7,056 datapoints. Almost 80% of the surveyed programs have some form of indexation, with about one-third of them doing so through automatic adjustments. Countries have evolved their approach to indexation significantly. The re- port’s 14 deep dives into specific country practices document that index- ation practices have also evolved remarkably over time, including in terms of altering methods, mechanisms, and frequency of indexation. While in- dexation is nearly a standard feature in higher-income contexts, a rich set of experiences is emerging across the income spectrum, including salient real-time developments in lower income contexts. Different types of indexation present comparative strengths and limitations. A system that adjusts transfers discretionarily may have more control over fiscal costs; but it also places those decisions on potentially less predict- able and objective – indeed discretionary – decision making processes. The politics of transfer augmentation is greatly reduced, but not eliminated, by automatic indexation; the predictability of automatic benefits yields sizable benefits, but the mechanics of constructing indexation measures also raises 59 06 — CONCLUSIONS KEEP THE PACE a set of data and technical challenges. In cases of skyrocketing inflation, the balance between maintaining purchasing power and fiscal sustainability should be carefully pondered. These considerations vary by country con- texts, with the level of maturity in ASP systems and the prevailing rate of inflation shaping appropriateness decisions significantly. Indexations should be interpreted within a wider set of macro and micro issues. For instance, a fiscal policy perspective should be closely in sync with monetary policy – a fusion that occurs, among others, in the con- text of unconventional monetary policy of large-scale cash transfer in- jections. The calibration of monetary and fiscal policy – and determining whether a cash injection would deter or foment inflation – are matters of macroeconomic debate. At micro level, mitigating inflation means consid- ering a wealth of options within social assistance. Among them, it is im- portant to dust off the traditional debate on cash versus in-kind.42 From this perspective, it is important not just to “index cash,” but to consider cost-benefit scenarios where it might be more effective and efficient to switch in transfer modalities. Identifying thresholds above which in-kind food, for example, is more appropriate than, say, vouchers or cash trans- fers is an important area of analytical inquire. The quandary of choosing transfer modalities that has now been reenergized in light of innovative de- livery practices that are blurring the lines between “cash,” “vouchers” and “in-kind food.” Electronic vouchers, time-bound cash transfers, the use of digital currencies and other options are making the spectrum of food as- sistance options both broader and more fluid. Ironing out and tailoring practical indexation choices. The selection of benchmark mechanisms between price, wage, or combinations thereof – as well as the relative weighting of those mechanisms – would entail the consideration of trade-offs between adequacy and fiscal costs in the short and longer-run. It has been documented that in normal times, nominal wages grow faster than prices: if countries establish a given budget envel- op, if indexation is based on wages and if the social assistance program is a medium-long term scheme, then there might be a trade-off between (lower) initial and (higher) subsequent adequacy; conversely, if anchored on prices, adequacy may decline over time relative to wages. As it was re- cently put, “adjustment to prices costs less than adjustment to wages”.43 This implies that initial level of indexation can be set at different thresh- 42 — Gentilini (2023, 2016) 43 — OECD (2022, p.5). 60 06 — CONCLUSIONS KEEP THE PACE olds pending on the mechanism of choice. Also, program goals and design matters: for example, social pensions programs (whether welfare-targeted or not) are devised to provide income support for seniors. This entails two decades of potential continued assistance. Such duration and objective may contrast with those of other programs meant to assist over the course of a narrower segment of the lifecycle (e.g., benefits for families with chil- dren aged 0-2), provide temporary countercyclical cushion, or offer short- term labor-intensive works. The above considerations on adequacy, pre- dictability, and costs in the short and longer-run may pan out differently in programs exhibiting fundamentally diverse goals and design parameters. Codifying such diversity operationally may be an important area for future applied research. 61 >> KEEP THE PACE REFERENCES MAIN TEXT AND ANNEXES Allen, J. and Gentilini, U. (2024) Cash Transfers and Inflation: An Overview of Evidence. World Bank and IFPRI. Washington, DC. Forthcoming. Balasundharam, V., Kayastha, A., and Poplawski-Ribeiro, M. (2023) Inflation Indexation in Public Fi- nances: A Global Dataset on Current Practices. IMF Working Papers 23/264. Barrett, C. (2013) Food Security and Sociopolitical Stability. (Ed). Oxford University Press. Bodewig, C., Calcutt, E., Conner, G., Cook, S., Gentilini, U., Hill, R., Majoka, Z., Narayan, A., Saidi, M., and Tassot, C. (2021) Stress Testing Social Protection: A Rapid Appraisal of the Adaptability of Social Protection Systems and Their Readiness to Scale-Up – A Guide for Practitioners. World Bank. Wash- ington, DC. Bowen, T., del Ninno, C., Andrews, C., Coll-Black, S., Gentilini, U., Johnson, K., Kawasoe, Y., Kryeziu, A., Maher, B., & Williams, A. (2020) Adaptive Social Protection: Building Resilience to Shocks. Interna- tional Development in Focus. World Bank, Washington, DC. CALP Network. (2021) Good Practice Review on Cash Assistance in Contexts of High Inflation and Depreciation. London. Cecchini, S. and Madariaga, A. (2011) Conditional Cash Transfer Programmes: The Recent Experi- ence in Latin America and the Caribbean. ECLAC. Santiago. Cheema, I., Farhat, M., Binci, M., Asmat, R., Javed, S., and O’Leary, S. (2020) Benazir Income Support Programme: Evaluation Report. Oxford Policy Management. Oxford. Devereux S. (1988) “Entitlements, Availability and Famine”. Food Policy 13(3): 270-282. DG ECHO. (2022) Cash Transfers. Thematic Policy Document 3. Luxembourg. FAOSTAT (2024) World Food Situation FAO Food Price Index. Online database. Gassmann, F., Gentilini, U., Morais, J., Nunnenmacher, C., Okamura, Y., Bordon, G., and Valleriani, G. (2023) “Is the Magic Happening? A Systematic Literature Review of the Economic Multiplier of Cash Transfers”. World Bank, Policy Research Working Paper 10529. Washington, DC. Gentilini, U. (2016) Revisiting the “Cash versus Food” Debate: New Evidence for an Old Puzzle?  The World Bank Research Observer, 31(1), 135–167. Gentilini, U. (2022) Cash Transfers in Pandemic Times. Evidence, Practices and Implications from the Largest Scale-up in History. World Bank. Washington, DC. Gentilini, U. (2023) Why does in-kind assistance persist when evidence favors cash transfers? Meet the six people shaping the debate. Brookings Commentary. Gentilini, U., Almenfi, M., Iyengar, H., Valleriani, G., Okamura, Y., Urteaga, E., Aziz, S., Noruzi, M. and Chu, M. (2023) “Tracking Global Social Protection Responses to Inflation – Living paper v.5”. World Bank, Social Protection and Jobs Discussion Paper 2305. Washington, DC. 62 >> KEEP THE PACE Gopinath, G. (2023) Gita Gopinath’s Introductory Remarks for the Conference “Fiscal Policy in an Era of High Debt”. Washington, DC. Government of Finland (2016) “Government’s proposal to parliament to amend Section 7 of the Child Benefit Act, HE 151/2016”. Helsinki. Government of Germany (2023) Citizen’s Benefits: Basic income support for jobseekers. Federal Ministry of Labor and Social Affairs. Bonn. Government of Ghana. (2023a) Indexation Mechanism of Benefits Under the Livelihood Empower- ment Against Poverty (LEAP) Cash Transfer Programme. Ministry of Finance. Accra. Government of Ghana (2023b) The Budget Statement and Economic Policy of the Government of Ghana for the 2023 Financial Year. Accra. Government of Lesotho (2024) Official Documents - Budget Formulation - Budget Speeches. Maseru. Government of Norway (2023) European Social Charter: Ad Hoc Report on the Cost-of-Living Crisis. RAP/RCha/NO/2023. Oslo. Hickey, S., T. Lavers, M. Niño-Zarazúa and Seekings, J. (2019) The Politics of Social Protection in East- ern and Southern Africa. Oxford University Press. ILO and UNICEF (2021) Mozambique Social Action Budget Brief 2021. Geneva and New York. Kagin, J., Qi, T., Kumar, D., Gupta, A., Taylor, E., Amondi, E., Clough, A., Gualtieri, A., Krishnaswamy, S., Leaduma, A., Monetta, C., Alvarado, W., Kyanjo, J., and Likicho, L. (2024) The Cost of Inaction: Im- pacts of WFP Refugee Assistance Shortfalls on Food Security Outcomes in Uganda. IDRA, UC Davis, Kagin’s Consulting, WFP, VT. Nairobi. Lagi, M., Bertrand, K. and Bar-Yam, Y. (2011) The food crises and political instability in North Africa and the Middle East. NECSI. Cambridge. Majoka, Z. (2024) “If you don’t move, you fall behind: Benefit indexation and the challenge of infla- tion”. World Bank. Islamabad and Washington, DC. McLean, C., Carraro, L., Barca, V., and Alfers, L. (2021) Transfer Values: How Much Is Enough? Balanc- ing social protection and humanitarian considerations. Social Protection Approaches to COVID-19 Expert Advice Service (SPACE). DAI. London. Nsabimana, A., Swain, R., Surry, Y. and Ngabitsinze, J. (2020) “Income and Food Engel Curves in Rwanda: A Household Microdata Analysis”. Agricultural and Food Economics 8(11) (doi.org/10.1186/ s40100-020-00154-4) OECD (2022) How Inflation Challenges Pensions. Paris. OECD (2023) Taxing Wages 2023: Indexation of Labour Taxation and Benefits in OECD Countries. Paris. OECD (2024) Modernising Access to Social Protection: Strategies, Technologies and Data Advances in OECD Countries. Paris. Perry, B. (2022) Child Poverty in New Zealand. Ministry of Social Development. Wellington. Sen A. (1981) Poverty and Famines. An Essay on Entitlements and Deprivation. Oxford University Press Tesliuc, C., Corral, P., Gupta, S. and Ampredu, C. (2024) Ghana Livelihood Empowerment Against Pov- erty (LEAP): Assessment of doubling LEAP benefit level in 2024. Accra. Mimeo. Timmer, P, Falcon, W., Pearson, S. (1983) Food Policy Analysis. World Bank. Washington, DC. UN CEPAL (undated). Base de datos de programas de protección social no contributiva en América Latina y el Caribe. Santiago. World Bank (2024). Food Security Update (May 30). Washington, DC. 63 >> KEEP THE PACE COUNTRY CASE STUDIES 01. Arimbi, H. R. (2024a). Belgium: Guaranteed Minimum Income Benefit for the Elderly (GRAPA). 02. Arimbi, H. R. (2024b). Canada – The Old Age Security Program (OAS). 03. Aziz, S. (2024a). The Citizen’s Benefit Program in Germany.  04. Aziz, S. (2024b). The Housing Benefit Program in Norway. 05. Cheng, C. (2024a). Australia’s Age Pension programme. 06. Cheng, C. (2024b). Moving from an automatic to ad hoc adjustment in a low inflation setting. A case study from Maldives’ Old Age Basic Pensions Scheme (BP). 07. Nogueira, J. L. M. (2024a). An analysis of index-linking and adequacy of the Chilean Securities and Opportunities Program. 08. Nogueira, J. L. M. (2024b). Brazil’s Bolsa Familia: A Case Study to use indexation to avoid politici- sation of a cash transfer program. 09. Nogueira, J. L. M. (2024c). India: Mahatma Gandhi National Rural Employment Guarantee Act. 10. TMM Iyengar, H. (2024). New Zealand’s attempt to close the gap in living standards between so- cial program beneficiaries and wage earners: A case study on the Indexation of Jobseeker Support program. 11. Trujillo, M. A. (2024a). Case study: Argentina Universal Child Allowance Program (AUH). 12. Trujillo, M. A. (2024b). Case study: Mexico Pension for the Well-being of Older People (PBPAM). 13. Trujillo, M. A. (2024c). Case study: Uruguay Family Allowances of the Equity Plan Calvin. 14. Valleriani, G. (2024). Incapacity Pension in Italy: Evolution of Indexation Practices and Benefits Adjustment. CASE STUDY REFERENCES Argentina 01. ANSES. (2023). Asignación Familiar por Hija e Hijo. https://www.anses.gob.ar/hijas-e-hijos/asig- nacion-familiar-por-hija-e-hijo#:~:text=Para%20garantizar%20los%20derechos%20a,y%20ni%C3%- B1o%20de%20la%20Argentina&text=A%20partir%20de%20marzo%202023,a%20ser%20de%20 %24404%20mil  02. ECLAC. (2023). Base de datos de programas de protección social no contributiva en América Latina y el Caribe. Asignación Universal por Hijo para Protección Social (2009) Argentina. https://dds. cepal.org/bpsnc/programa?id=162  03.IMF. (2023). World Economic Outlook April 2023: A Rocky Recovery. https://www.imf.org/en/Pub- lications/WEO/Issues/2023/04/11/world-economic-outlook-april-2023  04. Instituto Nacional de Estadística y Censos (INDEC). (2020). Encuesta Nacional de Gastos de los Hogares 2017-2018 Informe de ingresos. https://www.indec.gob.ar/ftp/cuadros/sociedad/eng- ho_2017_2018_informe_gastos.pdf  05. Instituto Nacional de Estadística y Censos (INDEC). (2023). Series de tiempo. Índice de precios al Consumidor (IPC). https://www.datos.gob.ar/series/api  64 >> KEEP THE PACE Australia 01. Australian Bureau of Statistics. (1988). History of pensions and other benefits in Australia. https:// www.abs.gov.au/ausstats/abs@.nsf/94713ad445ff1425ca25682000192af2/8e72c4526a94aaed- ca2569de00296978!opendocument  02.Australian Bureau of Statistics. (2023). Average weekly earnings, Australia. https://www.abs. gov.au/statistics/labour/earnings-and-working-conditions/average-weekly-earnings-australia/lat- est-release  03. Australian Bureau of Statistics. (2023). Selected living cost indexes, Australia. https://www.abs. gov.au/statistics/economy/price-indexes-and-inflation/selected-living-cost-indexes-australia/lat- est-release  04.Australian Department of Social Services. (2023). Age Pension. https://www.dss.gov.au/seniors/ benefits-payments/age-pension  05. Australian Department of Social Services. (2023). DSS Annual Report. https://www.dss.gov.au/ sites/default/files/documents/10_2023/dss-annual-report-published-version_0.pdf  06. Australian Department of Social Services. (2023). Social security guide. https://guides.dss.gov. au/social-security-guide/5/2/2/10  07.  Australian Department of Social Services. (n.d.). Annual Reports. https://www.dss.gov.au/ about-the-department/publications-articles/corporate-publications/annual-reports  08.Michelle Grattan. (2023, 22 August). Age pension cost to ease by 2060s but super tax breaks to swell: Intergenerational report. The Conversation. https://theconversation.com/age-pension-cost- to-ease-by-2060s-but-super-tax-breaks-to-swell-intergenerational-report-212012  09. Ministers for the Department of Social Services. (2023). Media Releases. https://ministers.dss. gov.au/media-releases/12366  10. OECD. (2021). Pensions at a Glance 2021: OECD and G20 Indicators. https://www.oecd.org/els/ public-pensions/PAG2021-country-profile-Australia.pdf  11. Parliamentary Library. (2014, 14 April). Pension Indexation: a Brief History. FlagPost. https://www. aph.gov.au/About_Parliament/Parliamentary_Departments/Parliamentary_Library/FlagPost/2014/ April/Pension-indexation  12. Services Australia. (2023). Assets test for Age Pension. https://www.servicesaustralia.gov.au/ assets-test-for-age-pension?context=22526  13. The Treasury Australian Government. (2001). Article 3 - Towards higher retirement incomes for Aus- tralians: a history of the Australian retirement income system since Federation. https://treasury.gov. au/publication/economic-roundup-centenary-edition-2001/article-3-towards-higher-retirement-in- comes-for-australians-a-history-of-the-australian-retirement-income-system-since-federation 14. The Treasury Australian Government. (2021). Sustainability Report. https://treasury.gov.au/sites/ default/files/2021-02/p2020-100554-ud04_sustainability.pdf  15. The Treasury Australian Government. (2023). 2023 Intergenerational Report. Australia’s future to 2063. https://treasury.gov.au/publication/2023-intergenerational-report  16. World Bank. (n.d.). Consumer price index (annual %). https://data.worldbank.org/indicator/FP.CPI. TOTL.ZG?locations=AU  17. World Bank. (n.d.). Population ages 65 and above (% of total population). https://data.worldbank. org/indicator/SP.POP.65UP.TO.ZS?locations=AU    65 >> KEEP THE PACE Belgium 01. Federal Public Service – Social Security. (updated 2024). Everything you always wanted to know about social security. https://socialsecurity.belgium.be/sites/default/files/content/docs/en/publi- cations/everything-social-security-en.pdf. 02. PensionStat.be. (n.d.). Garantie de revenus aux personnes âgées (GRAPA). https://www.pension- stat.be/fr/chiffres-cles/pension-legale/grapa 03.  Service Fédéral des Pensions. (n.d.). La garantie de revenus aux personnes âgées (GRAPA). https://www.sfpd.fgov.be/fr/droit-a-la-pension/grapa.   04. Statbel. (n.d.). Consumer Price Index. https://statbel.fgov.be 05. Statbel. (n.d.). Health Index. https://statbel.fgov.be   Brazil 01. Cotta, R., & Machado, J. (2013). The Bolsa Família cash transfer program and food and nutrition security in Brazil: a critical review of the literature. Revista Panamericana de Salud Pública 33(1):54-60. 02. De Souza, P., Guerreiro O. R., Paiva H. L., & Soares S. (2018). Os Efeitos do Programa Bolsa Família sobre a Pobreza e a Desigualdade: Um Balanço dos Primeiros Quinze Anos. Rio de Janeiro: Instituto de Pesquisa Econômica Aplicada (IPEA).    03. Gov.br. (2015). Bolsa Família complementa renda de 14 mi de famílias em janeiro. Ministério do Desenvolvimento Social e Combate à Fome. https://www.gov.br/casacivil/pt-br/assuntos/noti- cias/2015/janeiro/bolsa-familia-complementa-renda-de-14-mi-de-familias-em-janeiro.   04. Instituto Brasileiro de Geografia e Estatística. (n.d.). INPC - Índice Nacional de Preços ao Con- sumidor. https://www.ibge.gov.br/estatisticas/economicas/precos-e-custos/9258-indice-nacion- al-de-precos-ao-consumidor.html?=&t=o-que-e.    05. Lei Nº 10.836 de 9 de Janeiro de 2004. https://www.planalto.gov.br/ccivil_03/_ato2004- 2006/2004/lei/l10.836.htm.   Canada 01. Blackshaw, M., & Cahill, E. (2020). Research publications. Canada’s Retirement Income System. https://lop.parl.ca/sites/PublicWebsite/default/en_CA/ResearchPublications/201940E  02. Government of Canada. (2024). Eligibility criteria of OAS, GIS, and Allowance and Indexation Mechanism. https://www.canada.ca/en/services/benefits/publicpensions/cpp/old-age-security/ payments.html  03.  Government of Canada. (n.d.). Old Age Security (OAS) - Table of Benefit Amounts by marital status and income level. https://open.canada.ca/data/en/dataset/dfa4daf1-669e-4514-82cd- 982f27707ed0  04. Government of Canada. (n.d.). Statistics related to the Old Age Security program and the Cana- da Pension Plan. https://www.canada.ca/en/employment-social-development/programs/pensions/ pension.html  05.  Michaud, C. (2019). Development of a Consumer Price Index for Seniors. Statistics Canada.  https://www150.statcan.gc.ca/n1/pub/62f0014m/62f0014m2019004-eng.htm  06.  Office of the Superintendent of Financial Institutions Canada. (2021). Actuarial Report (18th) on the Old Age Security Program as at 31 December 2021. Ottawa, Ontario: Office of the Chief Actuary.  07. Statbel. (n.d.). Consumer Price Index. https://statbel.fgov.be.   66 >> KEEP THE PACE 08. Statistics Canada (2024). Upper-income limit, income share and an average of the adjusted mar- ket, total and after-tax income by income decile. https://www150.statcan.gc.ca/t1/tbl1/en/cv.ac- tion?pid=1110019301    Chile 01. Farías, C. (2012). Social Protection Systems in Latin American and the Caribbean – Chile. Santia- go, Chile: United Nations. https://repositorio.cepal.org/server/api/core/bitstreams/4c78b1e5-4ce4- 441a-a33d-ebe14e0605e5/content 02. Instituto Nacional de Estadística. (n.d.). Índice de Precios al Consumidor.  https://www.ine.gob.cl/ estadisticas/economia/indices-de-precio-e-inflacion/indice-de-precios-al-consumidor  03. Ley 20595: Crea el Ingreso Ético Familiar que establece bonos y transferencias condicionadas para las familias de pobreza extrema y crea subsidio al empleo de la mujer (2012). Ministerio de De- sarrollo Social, Subsecretaría de Evaluación Social. Bilioteca del Congreso Nacional de Chile. https:// www.bcn.cl/leychile/navegar?idNorma=1040157   04. United Nations - Economic Commission for Latin America and the Caribbean (ECLAC). (n.d.). Chile Securities and Opportunities - SSyOO (Ethical Family Income - IEF) (2012-). https://dds.cepal.org/ bpsnc/programme?id=61  05. World Bank. (n.d.). GDP (Current US$) - Chile. World Bank national accounts data, and OECD Na- tional Accounts data files. https://data.worldbank.org/indicator/NY.GDP.MKTP.CD?end=2022&loca- tions=CL&start=2012   Germany 01. Bruckmeier, K., Dummert, S., Grunau, P., Hohmeyer, K., & Lietzmann, T. (2020). New administra- tive data on Welfare Dynamics in Germany: The sample of Integrated Welfare Benefit Biographies (SIG). Journal for Labour Market Research, 54(1). https://doi.org/10.1186/s12651-020-00280-y 02. Bürgergeld Regelsatz 2024 so hoch ist der regelbedarf. Bürgergeld. (2024, January 28). https:// www.hartziv.org/regelsatz/ 03. Gavarini, L. (2024, February 7).  Bürgergeld: Wer bezieht es? Und wie viel Prozent Sind Deut- sche  Statsbürger?. SÜDKURIER. https://www.suedkurier.de/ueberregional/wirtschaft/geld-fi- nanzen/buergergeld-statistik-wer-bezieht-es-6-2-24;art1373668,11757767 04.Jacobi, L., & Kluve, J. (2006). Before and After the Hartz Reforms: The Performance of Active La- bour Market Policy in Germany. IZA Discussion Paper Series No. 2100. Bonn, Germany: IZA. https:// docs.iza.org/dp2100.pdf 05. Martholt, S. (2023, December 22). Bürgergeld Tabelle 2024. Bürgergeld. https://www-buerger- -geld-org.translate.goog/regelsatz/buergergeld-tabelle-2024/?_x_tr_sl=de&_x_tr_tl=en&_x_tr_ hl=en&_x_tr_pto=sc 06. OECD tax-benefit database. (n.d.). https://www.oecd.org/els/soc/taxben-germany-latest.pdf 07. Piekarz, von P. (2024, February 6). Bürgergeld formulare, Antrag 2024. Bürgergeld. https://www. buergergeld.org/formulare/ 08. Statistisches Bundesamt. (2021, June 7). Genesis-online. https://www.destatis.de/DE/Themen/ Wirtschaft/Preise/Verbraucherpreisindex/datenbank-teaser.html 09. The Local Germany. (2023, August 30).  Bürgergeld: Germany’s monthly unemployment benefit to rise by 12 percent. The Local. https://www.thelocal.de/20230830/burgergeld-germanys-month- ly-unemployment-benefit-to-rise-by-12-percent 67 >> KEEP THE PACE 10.Veröffentlicht von, M. M. (2024, January 31).  Empfänger von Alg II und Sozialgeld / Bürgergeld bis 2024. Statista. https://de.statista.com/statistik/daten/studie/242062/umfrage/leistungsemp- faenger-von-arbeitslosengeld-ii-und-sozialgeld/   India 01. Bacil, F., & Bird, N. (2021). Social Protection Profile: India. International Policy Centre for Inclusive Growth (IPC-IG). https://ipcig.org/sites/default/files/pub/en/OP490_Social_protection_profile_India.pdf   02. Computer Centre, Ministry of Statistics & Programme Implementation, Government of India. (n.d.). CPI-Rural Inflation Rate Series. 03. MGNREGA Management Information System. Managed by the Ministry of Rural Development, Gov- ernment of India. Designed, Developed, and Hosted by the National Informatics Centre, Government of India. Data on the MGNREGA program. https://nreganarep.nic.in/netnrega/MISreport4.aspx.   04. Ministry of Labour and Employment, Government of India. (n.d.). CPI (Agriculture Labourers), point-to-point rate of inflation (%). https://labourbureau.gov.in/rate-of-inflation-archive (until 2016) and https://labourbureau.gov.in/rate-of-inflation-cpi-al-rl (2017-2023).   05. Ministry of Law and Justice (Legislative Department). (2005). The National Rural Employment Guarantee Act, 2005. The Gazette of India Extraordinary.  https://rural.gov.in/sites/default/files/nre- ga/Library/Books/1_MGNREGA_A.   06. Ministry of Rural Development. (2007). Annual Report 2006-2007. https://rural.gov.in/en/publi- cations/annual-report.  07. Ministry of Rural Development. (2008). Annual Report 2007-2008. https://rural.gov.in/en/publi- cations/annual-report.  08. Ministry of Rural Development. (2009). Annual Report 2008-2009. https://rural.gov.in/en/publi- cations/annual-report.  09. Ministry of Rural Development. (2010). Annual Report 2009-2010. https://rural.gov.in/en/publi- cations/annual-report.  10. Ministry of Rural Development. (2011). Annual Report 2010-2011. https://rural.gov.in/en/publica- tions/annual-report.  11. Ministry of Rural Development. (2012). Annual Report 2011-2012. https://rural.gov.in/en/publica- tions/annual-report.  12. Ministry of Rural Development. (2013). Annual Report 2012-2013. https://rural.gov.in/en/publica- tions/annual-report.  13. Ministry of Rural Development. (2015). Annual Report 2014-2015. https://rural.gov.in/en/publica- tions/annual-report.  14. Ministry of Rural Development. (2016). Annual Report 2015-2016. https://rural.gov.in/en/publica- tions/annual-report.  15. Ministry of Rural Development. (2017). Annual Report 2016-2017. https://rural.gov.in/en/publica- tions/annual-report.  16. Ministry of Rural Development. (2018). Annual Report 2017-2018. https://rural.gov.in/en/publica- tions/annual-report.  17. Ministry of Rural Development. (2019). Annual Report 2018-2019. https://rural.gov.in/en/publica- tions/annual-report.  18. Ministry of Rural Development. (2020). Annual Report 2019-2020. https://rural.gov.in/en/publi- cations/annual-report.  68 >> KEEP THE PACE 19. Ministry of Rural Development. (2021). Annual Report 2020-2021. https://rural.gov.in/en/publica- tions/annual-report.  20. Ministry of Rural Development. (2022). Annual Report 2021-2022. https://rural.gov.in/en/publi- cations/annual-report.  21. Ministry of Rural Development. (2023). Annual Report 2022-2023. https://rural.gov.in/en/publi- cations/annual-report.  22. Ministry of Rural Development, MGNREGA Division. (2017). Report of the Committee on Alignment of MGNREGA Wages with Minimum Agricultural Wages.    23. World Bank. (n.d.). India GDP. https://data.worldbank.org/indicator/NY.GDP.MKTP. CD?end=2022&locations=IN&most_recent_year_desc=true&start=2005.     Italy 01. Agenzia Iura (2023, September 4). Pensioni di Invalidità e Limiti Reddituali: Rettifica di INPS. https://www.agenziaiura.it/2023/09/04/pensioni-di-invalidita-e-limiti-reddituali-rettifica-di-in - ps/#:~:text=Per%20le%20pensioni%2C%20il%20limite,%2C88%20euro%20l’anno  02. Benetti, M. (2019, February). Legge di Bilancio e Perequazione. Barometro Cisl n. 12. https://www. cisl.it/wp-content/uploads/2019/05/Perequazione%20pensioni%20-%20storia.pdf  03. CGIL. (2021). Guida alla Lettura della Busta Paga. https://www.filcams.cgil.it/article/guide_sca- dute/breve_guida_sulla_busta_paga  04. Diversabili. (2018, December 17). Assegno Sociale: dal 2019 Serviranno 67 Anni. Novità Anche per gli Invalidi Civili. https://diversabili.it/2018/12/assegno-sociale-dal-2019-serviranno-67-anni-novi- ta-anche-per-gli-invalidi-civili/   05. Fisco e Tasse. (2022, December 18). Rivalutazione Pensioni 2023: gli Aumenti Previsti. https://www. fiscoetasse.com/rassegna-stampa/33201-rivalutazione-pensioni-2023-gli-aumenti-previsti.html   06.  Fisco e Tasse. (2023, December 1). Aumento Pensioni 2024: Ecco Come Calcolare gli Importi. https://www.fiscoetasse.com/approfondimenti/15743-aumento-pensioni-2024-ecco-come-cal- colare-gli-importi.html   07. INPS. (2022). Che Cos’è la Perequazione?. https://servizi2.inps.it/docallegati/Mig/AllegatiNews/ Perequazione_InpsCard_pdf.pdf  08. INPS. (n.d.). Osservatorio Statistico sulle Pensioni Erogate da INPS. https://servizi2.inps.it/ servizi/osservatoristatistici/api/getAllegato/?idAllegato=1036  09. INPS. (2023). Pensione di Inabilità agli Invalidi Civili. https://www.inps.it/it/it/dettaglio-appro- fondimento.schede-informative.pensione-di-inabilit-agli-invalidi-civili.html#:~:text=Il%20benefi- cio%20è%20corrisposto%20agli,in%20forma%20stabile%20in%20Italia  10.  IPSOA. (2023, December 18). Perequazione delle Pensioni: Cosa Cambia con l’Anticipo del Conguaglio al 1° Dicembre. https://www.ipsoa.it/documents/quotidiano/2023/12/18/perequazi- one-pensioni-cambia-anticipo-conguaglio-1-dicembre  11. ISTAT. (2018). Le Pensioni di Invalidità Civile nel Periodo 1980-97. https://www.istat.it/it/ files/2018/07/2000_6-1.pdf  12. ISTAT. (2023a). Domande Frequenti sui Prezzi al Consumo. https://www.istat.it/it/ar- chivio/16333#:~:text=Per%20gli%20indici%20NIC%20e,finale%20arrotondato%20a%201%20deci - male.&text=il%20risultato%20finale%20è%20arrotondato%20a%201%20decimale  13. ISTAT. (2023b). Il Sistema dei Prezzi al Consumo. https://www.istat.it/it/archivio/17484#:~:tex- t=Per%20gli%20organi%20di%20governo,un%20lavoratore%20dipendente%20  69 >> KEEP THE PACE 14.  Italian Government. (2023). Perequazione Automatica. https://temi.camera.it/leg19/post/fo- cus-perequazione-automatica-1.html#:~:text=Lgs.,%2C%20utilizzando%20l’indice%20FOI  15.  Law 3 June 1975, n. 160. https://www.normattiva.it/uri-res/N2Ls?urn:nir:stato:leg- ge:1975-06-03;160~art7  16.  Law 30 April 1969, n. 153. https://www.normattiva.it/uri-res/N2Ls?urn:nir:stato:leg- ge:1969-04-30;153~art19  17. Ministry of Labor and Social Policy. (2008, October). Rapporto di monitoraggio sulle politiche so- ciali. https://www.lavoro.gov.it/sites/default/files/archivio-doc-pregressi/AreaSociale_FondoNazi- onale/Aggiornamento.pdf  18. Pensioni&Lavoro. (n.d.). Guida alla Previdenza Obbligatoria. https://www.pensionielavoro.it/site/ home/guida-alla-previdenza-obbligatoria/quali-sono-le-principali-prestazioni-erogate/le-princi- pali-prestazioni-assistenziali.html  19. Rogari, M. (2023a). Pensioni, Conguaglio per l’Inflazione Anticipato al 1 Dicembre 2023. Il Sole 24 Ore. https://www.ilsole24ore.com/art/pensioni-scatta-1-dicembre-2023-conguaglio-rivalutazi- one-ecco-come-funziona-AFDinGIB?refresh_ce=1  20. Rogari, M. (2023b). Spesa di 231 miliardi per 17,7 milioni di pensioni Inps: un terzo sono anticipate, il 22,8% assistenziali. Il Sole 24 Ore. Il Sole 24 Ore. https://www.ilsole24ore.com/art/spesa-231-mil- iardi-le-177-milioni-pensioni-inps-terzo-sono-anticipate-228percento-assistenziali-AEqSxK8C  21. SISTAN. (2023, April 4). Le Pensioni nel Settore Privato. https://www.sistan.it/index.php?id=88&- no_cache=1&tx_ttnews%5Btt_news%5D=10775  22. SIULP. (2023, February 17). Perequazione Pensionistica dei Trattamenti Superiori a Quattro Volte il Minimo. https://siulp.it/perequazione-pensionistica-dei-trattamenti-superiori-a-quat- tro-volte-il-minimo/  23. 50ePiùEnasco. (2019, January 17). Trasformazione dell’invalidità civile in assegno sociale sos- titutivo. https://www.50epiuenasco.it/senza-categoria/trasformazione-invalidita-civile-in-asseg- no-sociale-sostitutivo/    Maldives 01.  Alvarenga, K. J., Franciscon, I., & Lazzarini, L. (2021). Social protection profile: Maldives. Inter- national Policy Centre for Inclusive Growth. https://ipcig.org/sites/default/files/pub/en/OP491_So- cial_protection_profile_Maldives.pdf  02. Asia Pacific Office of the United Nations Population Fund. (2022). Case study: Maldives. https:// asiapacific.unfpa.org/sites/default/files/pub-pdf/casestudy-maldives_final.pdf  03. Avas. (2023, September 22). Pres Solih pledges MVR 7k monthly allowance for the elderly. https:// avas.mv/en/135831  04. Davis, B., & Handa, S. (2015). Transfer size in selected national cash transfer programmes in sub-Sa- haran Africa. UNICEF. https://www.unicef-irc.org/publications/pdf/RESEARCH-BRIEF-2015-1.pdf  05. International Labour Organization. (n.d.). Social protection for older persons: Key policy trends and statistics. https://www.social-protection.org/gimi/gess/RessourcePDF.action?id=53957  06.  International Monetary Fund. (n.d.). DataMapper: Inflation rate, end of period consumer prices (Percent change) for Maldives. https://www.imf.org/external/datamapper/PCPIEPCH@WEO/MD- V?zoom=MDV&highlight=MDV 07. Maldives Pension Administration Office. (2009). Maldives Pension Act. https://pension.gov.mv/ storage/1463/Fsc74ky26wXfA6YihiQ0TON4tXFkIWAtRmt4tgEZ.pdf  70 >> KEEP THE PACE 08. Maldives Pension Administration Office. (2019). Consolidated Unofficial Translation of the Mal- dives Pension Act and Amendments. https://pension.gov.mv/storage/1457/mI9FIR242kUAkC9HuH- qmlybUejDgmobkGWudaTWW.pdf  09.  Statistics Maldives. (2021). Household Income and Expenditure Survey 2019. https://statis- ticsmaldives.gov.mv/nbs/wp-content/uploads/2021/05/HIES2019-Expenditure.pdf  10. United Nations Development Programme. (2021, May). Maldives-Bulletin. https://www.undp.org/ sites/g/files/zskgke326/files/migration/mv/Maldives-Bulletin-May-2021-Final.pdf  11. World Bank. (1994). Averting the old age crisis: Policies to protect the old and promote growth. Oxford University Press.    Mexico 01. CONAPO. (2023). Datos abiertos. Distribución territorial de la población. http://www.conapo.gob. mx/es/CONAPO/Datos_Abiertos_Distribucion_Territorial_de_la_Poblacion   02. CONEVAL. (2018). Informe de Evaluación de la Política de Desarrollo Social 2018. https://www. coneval.org.mx/evaluacion/iepsm/iepsm/documents/iepds_2018.pdf 03. CONEVAL. (2020). Ficha Inicial de Monitoreo 2019-2020 Pensión para el Bienestar de las Perso- nas Adultas Mayores. https://www.gob.mx/cms/uploads/attachment/file/601321/S176_PBAM.pdf    04. CONEVAL. (2022). Ficha de Monitoreo 2021-2022 Pensión para el Bienestar de las Personas Adul- tas Mayores. https://www.gob.mx/cms/uploads/attachment/file/778745/S176InformeFinal.pdf   05. CONEVAL. (2023). Evolución mensual del valor monetario de la Línea de Pobreza Extrema por Ingresos (Canasta básica alimentaria) enero 1992-noviembre 2023. https://www.coneval.org.mx/ Medicion/MP/Paginas/Lineas-de-Pobreza-por-Ingresos.aspx   06. CONEVAL. (2023). Preguntas Frecuentes. ¿Qué son las líneas de bienestar?. https://www.coneval. org.mx/quienessomos/Paginas/Preguntas-Frecuentes-CONEVAL.aspx   07. Diario Oficial de la Federación. (2003, September 25). Acuerdo por el que se emiten y publi- can las Reglas de Operación del programa de atención a los adultos mayores a cargo de la Sec- retaria de Desarrollo Social para el ejercicio fiscal 2003. https://dof.gob.mx/nota_detalle.php?codi- go=727068&fecha=25/09/2003#gsc.tab=0   08. DOF. (2004, September 30). ACUERDO por el que se modifican las Reglas de Operación del Pro- grama de Atención a los Adultos Mayores a cargo de la Secretaría de Desarrollo Social, publicadas el 25 de septiembre de 2003, para el Ejercicio Fiscal 2004.   https://www.dof.gob.mx/nota_detalle. php?codigo=660968&fecha=30/09/2004#gsc.tab=0   09. DOF. (2007, February 28). ACUERDO por el que se emiten y publican las Reglas de Operación del Programa de Atención a los Adultos Mayores de 70 años y más en zonas rurales, para el ejercicio fiscal 2007. https://www.dof.gob.mx/nota_detalle.php?codigo=4964186&fecha=28/02/2007#gsc.tab=0   10. DOF. (2013, December 29). ACUERDO por el que se emiten las Reglas de Operación del Programa de Pensión para Adultos Mayores, para el ejercicio fiscal 2014. https://www.dof.gob.mx/nota_detal- le.php?codigo=5328387&fecha=29/12/2013#gsc.tab=0    11. DOF. (2013, February 26). Acuerdo por el que se emiten las Reglas de Operación del Programa de Pensión para Adultos Mayores, para el ejercicio fiscal 2013. https://www.dof.gob.mx/nota_detalle. php?codigo=5288941&fecha=26/02/2013#gsc.tab=0  12. DOF. (2014, December 27). ACUERDO por el que se emiten las Reglas de Operación del Programa Pensión para Adultos Mayores, para el ejercicio fiscal 2015. https://www.dof.gob.mx/nota_detalle. php?codigo=5377505&fecha=27/12/2014#gsc.tab=0   71 >> KEEP THE PACE 13. DOF. (2014, December 27). ACUERDO por el que se emiten las Reglas de Operación del Programa Pensión para Adultos Mayores, para el ejercicio fiscal 2015. https://www.dof.gob.mx/nota_detalle. php?codigo=5377505&fecha=27/12/2014#gsc.tab=0   14. DOF. (2015, December). ACUERDO por el que se emiten las Reglas de Operación del Programa Pensión para Adultos Mayores, para el ejercicio fiscal 2016. https://www.dof.gob.mx/nota_detal- le_popup.php?codigo=5421758   15. DOF. (2016, December 28). ACUERDO por el que se emiten las Reglas de Operación del Programa Pensión para Adultos Mayores, para el ejercicio fiscal 2017. https://www.dof.gob.mx/nota_detalle. php?codigo=5467903&fecha=28/12/2016#gsc.tab=0   16. DOF. (2017, December 28). ACUERDO por el que se emiten las Reglas de Operación del Programa Pensión para Adultos Mayores, para el ejercicio fiscal 2018. https://www.dof.gob.mx/nota_detalle. php?codigo=5509626&fecha=28/12/2017#gsc.tab=0   17. DOF. (2019, December 31). Acuerdo por el que se emiten las Reglas de Operación del Programa Pensión para el Bienestar de las Personas Adultas Mayores, para el ejercicio fiscal 2020. https:// www.dof.gob.mx/nota_detalle.php?codigo=5583304&fecha=31/12/2019#gsc.tab=0    18. DOF. (2019, February 28). ACUERDO por el que se emiten las Reglas de Operación de la Pensión para el Bienestar de las Personas Adultas Mayores, para el ejercicio fiscal 2019. https://dof.gob.mx/ nota_detalle.php?codigo=5551445&fecha=28/02/2019   19. DOF. (2020, December 22). Acuerdo por el que se emiten las Reglas de Operación del Programa Pensión para el Bienestar de las Personas Adultas Mayores, para el ejercicio fiscal 2021. https:// www.dof.gob.mx/nota_detalle.php?codigo=5608440&fecha=22/12/2020#gsc.tab=0   20. DOF. (2020, May 08). DECRETO por el que se reforma y adiciona el artículo 4o. de la Constitución Política de los Estados Unidos Mexicanos. https://www.diputados.gob.mx/LeyesBiblio/proceso/do- cleg/64/242_DOF_08may20.pdf   21. DOF. (2021, July 07). Acuerdo por el que se modifica el diverso por el que se emiten las Reglas de Operación del Programa Pensión para el Bienestar de las Personas Adultas Mayores, para el ejercicio fiscal 2021, publicado el 22 de diciembre de 2020. https://www.dof.gob.mx/nota_detalle.php?codi- go=5623150&fecha=07/07/2021#gsc.tab=0   22. DOF. (2022, December 30). Acuerdo por el que se emiten las Reglas de Operación del Programa Pensión para el Bienestar de las Personas Adultas Mayores, para el ejercicio fiscal 2023. https://dof. gob.mx/nota_detalle.php?codigo=5676228&fecha=30/12/2022#gsc.ta&gsc.tab=0   23. DOF. (2021, December 30). Acuerdo por el que se emiten las Reglas de Operación del Programa Pensión para el Bienestar de las Personas Adultas Mayores, para el ejercicio fiscal 2022. https:// www.dof.gob.mx/nota_detalle.php?codigo=5639752&fecha=30/12/2021#%20gsc.tab=0&gsc.tab=0   24. Gobierno de México. (2023). Pensión para el Bienestar de las Personas Adultas Mayores. Apoya de manera universal a mujeres y hombres mayores de 65 años en todo el país. https://www.gob. mx/bienestar/acciones-y-programas/pension-para-el-bienestar-de-las-personas-adultas-may- ores-296817   25. INEGI. (2023). Índice Nacional de Precios al Consumidor (INPC). https://www.inegi.org.mx/temas/inpc/ 26.  International Monetary Fund. (2023). DataMapper: Inflation rate, end of period consumer prices (Percent change) for Mexico. https://www.imf.org/external/datamapper/PCPIEPCH@WEO/MDV/MEX   27. Secretaría de Bienestar. (2023, January 2). Pensión de personas adultas mayores aumenta a cuatro mil 800 pesos bimestrales. Comunicado 001.- Del 4 al 10 de enero se realizará la dispersión de pagos, correspondiente al bimestre enero-febrero. https://www.gob.mx/bienestar/prensa/pension-de-per- sonas-adultas-mayores-aumenta-a-cuatro-mil-800-pesos-bimestrales-ariadna-montiel   72 >> KEEP THE PACE 28. Secretaría de Hacienda y Crédito Público. (2023). Presupuesto de Egresos de la Federación 2019- 2023. https://www.finanzaspublicas.hacienda.gob.mx/es/Finanzas_Publicas/Paquete_Economi- co_y_Presupuesto   New Zealand 01. Fletcher, M. (2018). Welfare State: Repairs and redesign. Proceedings: Summit – Rethink- ing the Welfare System for the 21st Century. Wellington: Child Poverty Action Group Aotearoa New Zealand. https://static1.squarespace.com/static/60189fe639b6d67b861cf5c4/t/622ec3d- 05b8af2744511e385/1647231972628/Summit+Proceedings+2018+Final.pdf.  02. Mckenzie, A. (2023). Social Assistance Chronology 1844–2023: A historical summary of social se- curity benefits, war pensions, retirement pensions, taxation measures, family assistance, housing assistance, student support and labour market programmes. Ministry of Social Development, New Zealand. https://www.msd.govt.nz/documents/about-msd-and-our-work/about-msd/history/so- cial-assistance-chronology.docx  03. MSD. (2012a). Welfare Reform - Proposals for Bill Two of Welfare Reform. Regulatory Impact Statement. Ministry of Social Development, New Zealand.   https://www.msd.govt.nz/documents/ about-msd-and-our-work/publications-resources/regulatory-impact-statements/ris-welfare-re- form-phase-one-social-security-amendment-bill-no1-.pdf.  04. MSD. (2012b). Welfare Reform: Phase One – Social Security Amendment Bill (No. 1). Regulatory Impact Statement. Ministry of Social Development, New Zealand.  https://www.msd.govt.nz/docu- ments/about-msd-and-our-work/publications-resources/regulatory-impact-statements/ris-wel- fare-reform-detailed-proposals.pdf.  05. MSD. (2019a). Orders in Council and Amendment regulation for the Annual general Adjustment of rates and Thresholds. Cabinet papers - Information releases. Ministry of Social Development, New Zealand. https://www.msd.govt.nz/documents/about-msd-and-our-work/publications-resourc- es/information-releases/annual-general-adjustment-2019/cabinet-paper-annual-general-adjust- ment-of-rates-and-thresholds.pdf  06. MSD. (2019b). Social Assistance Legislation (Budget 2019 Welfare Package) Amendment Bill. Cabinet papers - Information releases. Ministry of Social Development, New Zealand. https://www. msd.govt.nz/about-msd-and-our-work/publications-resources/information-releases/cabinet-pa- pers/2019/social-assistance-legislation-budget-2019-welfare-package.html  07. MSD. (2020a). COVID-19 Recovery Package: increases to main benefits and the winter energy pay- ment. Cabinet papers - Information releases. Ministry of Social Development, New Zealand. https:// covid19.govt.nz/assets/Proactive-Releases/proactive-release/COVID-19-Recovery-Package-In- creases-to-Main-Benefits-and-the-Winter-Energy-Payment-Paper.pdf.   08. MSD. (2020b). Advice on different options for improving main benefits. Cabinet papers - Informa- tion releases. Ministry of Social Development, New Zealand. https://www.msd.govt.nz/documents/ about-msd-and-our-work/publications-resources/information-releases/cabinet-papers/2021/ budget/13-nov-2020-report-rep-20-11-1081-advice-on-different-options-for-increasing-main- benefits.pdf.  09. MSD. (2021). Budget 2021 Main Benefit Rate Increases and Related Regulatory Changes. Cabinet papers - Information releases. Ministry of Social Development, New Zealand. https://www.msd.govt. nz/about-msd-and-our-work/publications-resources/information-releases/cabinet-papers/2021/ budget-2021-main-benefit-rate-increases-and-related-regulatory-changes.html.  10. MSD. (2022). Annual General Adjustment 2022 and Related Regulatory Changes. Cabinet papers - Information releases. Ministry of Social Development, New Zealand. https://www.msd.govt.nz/ documents/about-msd-and-our-work/publications-resources/information-releases/cabinet-pa- pers/2022/annual-general-adjustment/paper-annual-general-adjustment-2022-and-related-regu- latory-changes.pdf  73 >> KEEP THE PACE 11. MSD. (2023a). Welfare assistance package to help low-income people meet the increasing cost of living. Cabinet papers - Information releases. Ministry of Social Development, New Zealand. https:// www.msd.govt.nz/documents/about-msd-and-our-work/publications-resources/information-re- leases/welfare-assistance-package-to-help-low-income-people-meet-the-increasing-cost-of-liv- ing/paper-welfare-assistance-package-to-help-low-income-people-meet-the-increasing-cost-of- living.pdf.  12. MSD. (2023b). Annual General Adjustment 2023 and related Regulatory Changes. Cabinet papers - Information releases. Ministry of Social Development, New Zealand. https://www.msd.govt.nz/ documents/about-msd-and-our-work/publications-resources/information-releases/welfare-as- sistance-package-to-help-low-income-people-meet-the-increasing-cost-of-living/paper-welfare- assistance-package-to-help-low-income-people-meet-the-increasing-cost-of-living.pdf.  13. MSD. (2023c). Benefit Fact Sheets Snapshot September 2023 Quarter. Ministry of Social Devel- opment, New Zealand. https://www.msd.govt.nz/documents/about-msd-and-our-work/publica- tions-resources/statistics/benefit/2023/benefit-fact-sheets-snapshot-september-2023.pdf.  14. MSD. Benefit Fact Sheets releases, published quarterly. Ministry of Social Development, New Zealand. https://www.msd.govt.nz/about-msd-and-our-work/publications-resources/statistics/ benefit/index.html#Impactof2013WelfareReformonbenefittrends6.  15. Perry, B. (2022). MSD Child Poverty Report 2022. Ministry of Social Development, New Zealand. https://www.msd.govt.nz/documents/about-msd-and-our-work/publications-resources/research/ child-poverty-in-nz/2022-child-poverty-report.pdf.  16. Social Security Act 1964 [61HA]: repealed, on 26 November 2018, by section 455(1) of the Social Security Act 2018 (2018 No 32). Ministry of Social Development, New Zealand. https://www.legisla- tion.govt.nz/act/public/1964/0136/latest/DLM3267056.html#DLM3267056 17. Social Security Act 2018 [452A] inserted, on 6 June 2019, by section 8 of the Social Assistance Legislation (Budget 2019 Welfare Package) Amendment Act 2019 (2019 No 23). https://www.legisla- tion.govt.nz/act/public/2018/0032/latest/LMS209600.html?search=sw_096be8ed81d874f8_Annu- al+General+Adjustment_25_se&p=1.  18. Statistics New Zealand. Database consisting of data on GDP, inflation (CPI) and Wages (AWE [FTES]). https://infoshare.stats.govt.nz/.  19. WEAG. (2019). Whakamana Tāngata: Restoring Dignity to Social Security in New Zealand. Welfare Expert Advisory Group. https://www.weag.govt.nz/assets/documents/WEAG-report/aed960c3ce/ WEAG-Report.pdf  20. WEAG. (2022, September 28). Welfare system: statistics. Welfare Expert Advisory Group. https:// www.weag.govt.nz/background/welfare-system-statistics/.  21.  Work and Income. Data on Jobseeker Support benefits. Government of New Zealand. https:// www.msd.govt.nz/documents/about-msd-and-our-work/publications-resources/statistics/bene- fit/2023/benefit-fact-sheets-snapshot-september-2023.pdf.  22.  Year-end Financial Statements of the Government of New Zealand. Treasury, Government of New Zealand. https://www.treasury.govt.nz/publications/financial-statements-government/ year-end-financial-statements.    Norway 01. Expert group appointed by the then Ministry of Local Government and Modernization. (2022). Bostøtten - opprydning og forankring. https://www.regjeringen.no/contentassets/9a60203ca2434e- 3c813e79ceb289582a/bostoetten_rapport.pdf 02. Husbanken. (2024). Beregning av Bostøtte. https://www-husbanken-no.translate.goog/person/ bostotte/beregning-av-bostotte/?_x_tr_sl=no&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=sc 74 >> KEEP THE PACE 03. Husbanken. (2024). Bostøtte på Engelsk Rules Relating to Housing Allowance. https://nedlasting. husbanken.no/Filer/9e6e.pdf 04. Nyhus E. K. (n.d.).  Husbanken. Store norske leksikon. https://snl-no.translate.goog/Hus- banken?_x_tr_sl=no&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=sc 05. OECD tax-benefit database. (2023). https://www.oecd.org/els/soc/TaxBEN-Norway-latest.pdf 06. Stord commune. (n.d.). Housing benefits. https://www.stord.kommune.no/housing-bene- fits.428232.5tdc456.tct.html?tklang=en     07. Tidens Krav. (2009, July 1).  6.500 får bostøtte her I fylket. Tidens Krav. https://www.tk.no/ny- heter/6-500-far-bostotte-her-i-fylket/s/1-113-4438435     Uruguay 01. Colafranceschi, M., & Vigorito, A. (2013). Uruguay: evaluación de las políticas de transferencias. La estrategia de inclusión y sus desafíos*. In R. Rofman, Hacia un Uruguay más Equitativo. Los Desafíos del Sistema de Protección Social. Montevideo: World Bank.  02. IMF. (2023, May 23). Uruguay: 2023 Article IV Consultation-Press Release; Staff Report; and Statement by the Executive Director for Uruguay. https://www.elibrary.imf.org/view/jour- nals/002/2023/178/article-A001-en.xml 03. Lavalleja, M., & Tenenbaum, V. (2020). Mecanismos de focalización de los programas de trans- ferencias monetarias en América Latina. El caso del Uruguay. Estudios y Perspectivas - Oficina de la CEPAL en Montevideo(43). https://repositorio.cepal.org/server/api/core/bitstreams/771d08b5- c937-4bc9-b62a-470b3ff7e283/content 04. Lavalleja, M., & Tenenbaum, V. (2022). Análisis y recomendaciones sobre el diseño de los pro- gramas de transferencias monetarias en el Uruguay. Documentos de Proyectos CEPAL.  05. Law N° 18.719, Presupuesto Nacional de Sueldos Gastos e Inversiones. Ejercicio 2010 - 2014. (2010). https://www.impo.com.uy/bases/leyes/18719-2010  06.  Law Nº 17.823, Código de la Niñez y la Adolescencia. Pub D.O. 14 set/004 - No 26586 (2004). https://www.oas.org/dil/esp/codigo_ninez_adolescencia_uruguay.pdf  07. Law No 18.227, Asignaciones Familiares. Pub. D.O. 9 ene/008 - No 27401 (2007). https://oig.cepal. org/sites/default/files/2007_ley18227_ury.pdf  08. Law Nº 19.924, Presupuesto Nacional 2020-2024. C/611/2020 - No 147 (2020). https://www. gub.uy/ministerio-economia-finanzas/politicas-y-gestion/ley-19924-presupuesto-nacion - al-2020-2024?hrt=1386  09. Ministerio de Desarrollo Social. (2023). Asignaciones Familiares-Plan de Equidad, Edición 2023- Documento de diseño. https://www.gub.uy/ministerio-desarrollo-social/sites/ministerio-desarrol- lo-social/files/documentos/publicaciones/DINTAD_2023_Asignaciones%20Familiares%20PE.pdf 10. Ministerio de Desarrollo Social. (2023). El Gasto Público Social en Uruguay. Informe de Estimación 2019-2021. https://www.gub.uy/ministerio-desarrollo-social/sites/ministerio-desarrollo-social/ files/documentos/publicaciones/GPS_2019_2021_final.pdf 11. Ministerio de Salud Pública & UNICEF. (2021). Consumo aparente de alimentos y bebidas en los hogares uruguayos. Una mirada a la realidad nacional y en hogares donde viven niños menores de 5 años. https://bibliotecaunicef.uy/opac_css/doc_num.php?explnum_id=299   12. Ministerio de Salud Pública & UNICEF. (2021). Consumo aparente de alimentos y bebidas en los hogares uruguayos. Una mirada a la realidad nacional y en hogares donde viven niños menores de 5 años. https://bibliotecaunicef.uy/opac_css/doc_num.php?explnum_id=299 75 >> KEEP THE PACE ANEXX 1 / TAXONOMY OF INDICES INDEXATION/ADJ. TYPE BENCHMARK BENCHMARK VARIATION Prices CPI National CPI Pensioner and Beneficiary Living Cost Index (PBLCI) (Smoothed) Health Index CPI excl. tobacco Consumer Price Index for Agricultural Labour (CPI-AL) Consumer prices for blue and white-collar families (FOI) Harmonised Index of Consumer Prices (HICP) Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W) CPI and adjustable reference unit (set by decree) [specific to Uruguay] Growth of living costs among low-income households Reference Social Indicator (RSI) [specific to Romania and based on inflation] Average price inflation Food prices Basic food basket Market food prices Locally available nutritionally balanced food basket Others Sub-indices concerning housing cost in rented and owner- occupied housing Average price development at federal level Wages Income Statutory minimum monthly wage First-grade starting salary Wages in the labor market of the [program’s] interested areas Average earnings Average earnings of rural and urban residents Net average ordinary time weekly earnings Average (net) wage Average wage index (e.g., Índice Medio de Salarios in Uruguay) Average development of the net wages and salaries per employee Median minimum income Prices and macroeconomic Prices and tax CPI and social tax revenues variables Wages and macroeconomic Income and tax Mobility Index (evolution of salaries and tax resources) variables Prices, Wages, and other GDP, prices, wages Budget of the State, economic situation and the increase of wages macroeconomic variables of employees Social benefits, prices, wages CPI, average earnings of rural and urban residents, and benefit amounts of other social security programmes 76 >> KEEP THE PACE ANEXX 2 / PROGRAM LEVEL INFO ON KEY FEATURES OF BENEFIT ADJUSTMENTS COUNTRY PROGRAM NAME ARE BENEFIT ADJUSTMENT BENCHMARK FREQUENCY INCOME SP CATEGORY SP SUBCATEGORY ADJUSTED? MECHANISM INDICATOR/ GROUP MECHANISM AFR AFR AFR AFR AFR AFR AFR AFR AFR Angola Kwenda 1. Yes 2. Ad hoc 1. Prices LMIC Cash transfers UCT Benin Municipalities and Communities Support for 2. No LMIC Cash transfers UCT Social Services Expansion: Unconditional Transfer Botswana Ipelegeng (self-reliance) 1. Yes 2. Ad hoc UMIC Public works Cash for work Burkina Faso Nahouri Cash Transfers Pilot Project (NCTPP) 2. No LIC Cash transfers UCT; CCT Burundi Cash for Jobs Project (World Bank Project) 2. No LIC Public works Cash for work Cameroon Program 559: National solidarity and social 3. Not available LMIC Cash transfers UCT justice Cape Verde National Center for Social Pensions 1. Yes 2. Ad hoc LMIC Social pensions Old age social (non-contributory) pensions; Disability pensions Central African The Service Delivery and Support to Communi- 2. No LIC Cash transfers UCT Republic ties Affected by Displacement Project* Chad Programs for food security 3. Not available LIC Cash transfers UCT Comoros Productive safety net (Filets sociaux 3. Not available LMIC Public works Cash for work productifs) Congo, Democratic Eastern Recovery Project (P145196) 3. Not available LIC Public works Cash for work Republic of Congo, Republic of Lisungi project 2. No LMIC Cash transfers UCT; CCT Congo, Republic of FSA project 3. Not available LMIC Cash transfers CCT Côte d’Ivoire National Productive Social Nets Program 1. Yes 2. Ad hoc 5. Discretionary LMIC Cash transfers UCT Eswatini Old Age Grant 1. Yes 2. Ad hoc 1. Prices 5. Discretionary LMIC Social pensions Old age social (non-contributory) pensions Ethiopia Productive Safety Net 1. Yes 2. Ad hoc 1. Prices 1. Annual LIC Cash transfers; UCT; Cash for work Public works Gambia, The Post-crising response to food-nutrition 3. Not available 2. Wages LIC Public works Cash for work insecurity Ghana Livelihood Empowerment Against Poverty 1. Yes 1. Automatic 1. Prices 1. Annual LMIC Cash transfers UCT (LEAP) Guinea Labour-Intensive Public Works Programme 1. Yes 2. Ad hoc 2. Wages LIC Public works Cash for work Guinea-Bissau Program for the handicapped 3. Not available LIC Social pensions Disability pensions (non-contributory) Kenya Cash transfer for OVC (CT-OVC) 1. Yes 2. Ad hoc 1. Prices 5. Discretionary LMIC Cash transfers UCT Lesotho Old age pension 1. Yes 2. Ad hoc 1. Annual LMIC Social pensions Old age social (non-contributory) pensions Liberia Social Cash Transfer 1. Yes 2. Ad hoc 1. Prices 4. Other LIC Cash transfers UCT Madagascar Tosika Fameno 2. No LIC Cash transfers UCT Malawi Food and Cash Transfers (FACT) 1. Yes 1. Automatic 6. Monthly LIC Cash transfers UCT Mali Emergency Safety Nets Project (Jigisemejiri) 1. Yes 2. Ad hoc 5. Discretionary LIC Cash transfers UCT Mauritania Tekavoul 1. Yes 2. Ad hoc LMIC Cash transfers CCT Mauritius Basic Retirement Pension (BRP) zero pillar 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Social pensions Old age social retirement only* (non-contributory) pensions Mozambique Basic Social Subsidy Programme 1. Yes 2. Ad hoc 1. Annual LIC Cash transfers UCT Namibia Child grant 1. Yes 2. Ad hoc UMIC Cash transfers UCT Niger Social Safety Nets Project 2. No LIC Cash transfers UCT Nigeria FADAMA 3. Not available LMIC Cash transfers UCT Rwanda Vision 2020 Umurenge 3. Not available LIC Cash transfers; UCT; Cash for work Public works CCT Sao Tome and Needy Mothers (Mães Carenciadas) 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary LMIC Cash transfers CCT Principe Senegal National cash transfer programme 1. Yes 2. Ad hoc 5. Discretionary LMIC Cash transfers Old age social Seychelles Retirement Pension (RP) 1. Yes 1. Automatic 1. Prices HIC Social pensions pensions (non-contributory) Seychelles Social Welfare Assistance (SWA) 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary HIC Cash transfers UCT Sierra Leone Social Safety Nets Project (EP Fet Po) 1. Yes 2. Ad hoc 3. ad hoc LIC Cash transfers UCT South Africa Child Support Grant 1. Yes 2. Ad hoc 1. Annual UMIC Cash transfers UCT South Sudan Juba urban poor cash response pilot 3. Not available LIC Cash transfers UCT Sudan Zakat 3. Not available LIC Cash transfers UCT Tanzania Zanzibar’s Universal Social Pension 2. No LMIC Social pensions Old age social (non-contributory) pensions Togo CCT with conditions on nutrition (This should 3. Not available LIC Cash transfers CCT be the program ASPIRE refers to: Cash Transfer Program for Vulnerable Children in Northern Togo (P144484)) Uganda Nothern Uganda Social Action Fund (II) 3. Not available LIC Public works Cash for work Zambia Social Cash Transfer Scheme (SCT) 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary LIC Cash transfers UCT Zimbabwe Harmonised Social cash transfer (HSCT)* 2. No LMIC Cash transfers UCT 77 >> KEEP THE PACE EAP EAP EAP EAP EAP EAP EAP EAP EAP Australia Age pension 1. Yes 1. Automatic 1. Prices 2. Semi-annual HIC Social pensions Old age social (non-contributory) pensions Cambodia NOURISH project 3. Not available LMIC Cash transfers CCT China Old-age pension (Pension schemes for 1. Yes 1. Automatic 9. Prices, 9. Prices, UMIC Social pensions Old age social rural and non-salaried urban residents [non Wages, and Wages, and (non-contributory) pensions contirbutory]) other factors other factors China Dibao 1. Yes 1. Automatic 1. Prices1 1. Annual UMIC Cash transfers UCT Fiji Social Welfare pension 1. Yes 2. Ad hoc UMIC Social pensions Old age social (non-contributory) pensions Fiji Poverty Benefit Scheme (now called as Family 1. Yes 2. Ad hoc UMIC Cash transfers UCT assistance scheme) Indonesia Family Hope Program 1. Yes 2. Ad hoc 5. Discretionary LMIC Cash transfers CCT Japan Public Assistance Program 1. Yes 2. Ad hoc 1. Prices HIC Cash transfers UCT Kiribati Elderly Fund Pension 1. Yes 2. Ad hoc 3. ad hoc LMIC Social pensions Old age social (non-contributory) pensions Korea National Basic Livelihood Security Act 1. Yes 2. Ad hoc 1. Prices 1. Annual HIC Cash transfers UCT Malaysia Financial Assistance for the People of Malaysia 1. Yes 2. Ad hoc UMIC Cash transfers UCT (Bantuan Rakyat 1 Malaysia) Mongolia The Child Money programme 1. Yes 2. Ad hoc LMIC Cash transfers CCT Myanmar Maternal and Child Cash Transfer 1. Yes 2. Ad hoc LMIC Cash transfers CCT New Zealand Best Start tax credit (social assistance) 1. Yes 1. Automatic 1. Prices 7. Above a spe- HIC Cash transfers UCT cific threshold New Zealand Sole parent support (social assistance) 1. Yes 1. Automatic 2. Wages 1. Annual HIC Cash transfers CCT New Zealand Job Seeker Support Program 1. Yes 1. Automatic 2. Wages 1. Annual HIC Cash transfers CCT Papua New Guinea New Ireland Disability Benefit 1. Yes 2. Ad hoc LMIC Social pensions Old age social (non-contributory) pensions; Disability pensions Philippines Pantawid Pamilyang Pilipino Program (4Ps) 1. Yes 2. Ad hoc 5. Discretionary LMIC Cash transfers CCT Samoa Senior Citizens Benefit 1. Yes 2. Ad hoc LMIC Social pensions Old age social (non-contributory) pensions Solomon Islands Rapid Employment Program 3. Not available LMIC Public works Cash for work Thailand Old Age Allowance 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary UMIC Social pensions Old age social (non-contributory) pensions Thailand Welfare Card Program 2. No 2. Ad hoc UMIC Cash transfers UCT Tonga Social Welfare Scheme for the Elderly 1. Yes 2. Ad hoc 5. Discretionary UMIC Social pensions Old age social (non-contributory) pensions Vietnam Social allowances 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary LMIC Social pensions Old age social (non-contributory) pensions; Disability pensions Vietnam Subsidies for Tet holiday expenditure for poor 3. Not available LMIC Cash transfers UCT households ECA ECA ECA ECA ECA ECA ECA ECA ECA Albania Ndihma Ekonomike (full & partial benefits) 1. Yes 2. Ad hoc UMIC Cash transfers UCT Armenia Family Poverty Benefit (PMT), incl. one time/ 1. Yes 2. Ad hoc 3. ad hoc UMIC Cash transfers UCT lump-sum monetary assistance Austria Minimum income/Social Assistance (Be- 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers UCT darfsorientierte Mindestsicherung/ Sozialhilfe) Azerbaijan Targeted State Social Assistance (TSSA) 1. Yes 2. Ad hoc 3. ad hoc 1. Annual UMIC Cash transfers UCT Belarus Public Targeted Social Assistance (GASP) - 1. Yes 2. Ad hoc 3. ad hoc 1. Annual UMIC Cash transfers UCT benefit to purchase technical equipment for social rehabilitation Belarus Public Targeted Social Assistance (GASP) - 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Cash transfers UCT monthly social benefit Belarus Public Targeted Social Assistance (GASP) - 1. Yes 2. Ad hoc 3. ad hoc 1. Annual UMIC Cash transfers UCT one-time social benefit* Belgium Child Benefits 1. Yes 1. Automatic 1. Prices 6. Monthly HIC Cash transfers UCT Belgium Guaranteed Minimum Income Benefit for the 1. Yes 1. Automatic 1. Prices 6. Monthly HIC Social pensions Old age social Elderly (GRAPA) (non-contributory) pensions Bosnia and Herze- Assistance to Families with Children 1. Yes 2. Ad hoc UMIC Cash transfers UCT govina Bulgaria Monthly Social Assistance in accordance with 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary UMIC Cash transfers UCT the Social Assistance Act (GMI) (persons and families) Bulgaria One-time assistance 3. Not available UMIC Cash transfers UCT Czech Republic Parental allowance 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary HIC Cash transfers UCT Denmark Public pension scheme Folkepension (non 1. Yes 1. Automatic 2. Wages 1. Annual HIC Social pensions Old age social contributory) (non-contributory) pensions Estonia Pension Program (Sotsiaalkindlustusamet) 1. Yes 1. Automatic 7. Prices 1. Annual HIC Social pensions Old age social and Macro- (non-contributory) pensions; Disability economic pensions variables Finland Child benefit (Lapsilisä) 1. Yes 2. Ad hoc HIC Cash transfers UCT France Active solidarity income (RSA) 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers UCT Georgia Allowances for contracted doctors and nurses 1. Yes 2. Ad hoc UMIC Cash transfers UCT residing in highmountinous settlements Georgia Benefit for refugees and IDP 1. Yes 2. Ad hoc 5. Discretionary UMIC Cash transfers UCT Georgia Demographic situation improvement program 1. Yes 2. Ad hoc UMIC Cash transfers UCT 78 >> KEEP THE PACE Georgia Foster Care (Subprogram of the Social Reha- 1. Yes 2. Ad hoc UMIC Cash transfers UCT bilitation and Child care Program) Georgia Social Package-Political persecuted 1. Yes 2. Ad hoc 3. ad hoc 2. Semi-annual UMIC Cash transfers UCT Georgia Social Package-Survivors pension 1. Yes 2. Ad hoc 5. Discretionary UMIC Social pensions Disability pensions (non-contributory) Georgia State compensation and state stipend (police, 1. Yes 2. Ad hoc UMIC Cash transfers UCT military, and politicians) Georgia Targeted Social Assistance 1. Yes 2. Ad hoc 3. ad hoc UMIC Cash transfers UCT Germany Citizen Benefit (Bürgergeld) - formerly known 1. Yes 1. Automatic 5. Prices and 1. Annual HIC Cash transfers UCT as Hartz IV (Unemployment Benefits) Wages Greece Social Solidarity Income (SSI) 3. Not available HIC Cash transfers UCT Hungary Family allowance (Családi pótlék) 1. Yes 2. Ad hoc 5. Discretionary HIC Cash transfers CCT Iceland Family benefit (Barnabætur) 1. Yes 2. Ad hoc 3. ad hoc 1. Annual HIC Cash transfers UCT Iceland Municipality financial assistance (Fjárhag- 1. Yes 2. Ad hoc 3. ad hoc 1. Annual HIC Cash transfers UCT saðstoð sveitarfélaga) Ireland State Pension (Non-Contributory) 1. Yes 2. Ad hoc HIC Social pensions Old age social (non-contributory) pensions Italy Incapacity Pension (Pensione di Incapacità) 1. Yes 1. Automatic 1. Prices 1. Annual HIC Social pensions Disability pensions (non-contributory) Kazakhstan Targeted Social Assistance (TSA)* 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Cash transfers UCT Kosovo Social Assistance Scheme 1. Yes 2. Ad hoc 5. Discretionary UMIC Cash transfers UCT Kyrgyz Republic Monthly allowance for low-income families 1. Yes 2. Ad hoc 5. Discretionary LMIC Cash transfers UCT with children Latvia Childcare benefit (Bērna kopšanas pabalsts) 1. Yes 2. Ad hoc 8. Prices, Wages, and Macro-eco- HIC Cash transfers UCT nomic variables Latvia Family state benefit (Ģimenes valsts pabalsts) 1. Yes 2. Ad hoc 8. Prices, Wages, and Macro-eco- HIC Cash transfers UCT nomic variables Latvia Guaranteed minimum income benefit (Ga- 1. Yes 2. Ad hoc 2. Wages 1. Annual HIC Cash transfers UCT rantētā minimālā ienākuma pabalsts) Lithuania Child benefit (išmoka vaikui) 1. Yes 2. Ad hoc 1. Prices 1. Annual HIC Cash transfers UCT Luxembourg Child Benefit Program - The Children’s Future 1. Yes 1. Automatic 1. Prices 7. Above a spe- HIC Cash transfers UCT Fund (Caisse pour l’Avenir des Enfants) cific threshold Macedonia Child allowance (for recipients of SFA)* 3. Not available 1. Automatic 1. Prices 1. Annual UMIC Cash transfers UCT Macedonia Financial assistance to orphans (18-26) 1. Yes 3. Not available UMIC Cash transfers UCT Macedonia Foster families* 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Cash transfers UCT Macedonia Guaranteed Minimum Assistance (GMA)* 1. Yes 1. Automatic 1. Prices 1. Prices UMIC Cash transfers UCT Macedonia Parent Allowance* 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Cash transfers UCT Macedonia Parent Allowance for the 4th child* 1. Yes 1. Automatic 1. Prices 1. Prices UMIC Cash transfers UCT Moldova Ajutor Social 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Cash transfers UCT Netherlands Social assistance (Participatiewet) 1. Yes 1. Automatic 2. Wages 2. Semi-annual HIC Cash transfers UCT Norway Advance payments of child maintenance for 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers UCT lone parents (bidragsforskott) Norway Cash benefit for families with small children 1. Yes 2. Ad hoc 3. ad hoc 1. Annual HIC Cash transfers UCT (kontantstøtte) Norway Child benefit incl. lone-parent supplements 1. Yes 2. Ad hoc 1. Annual HIC Cash transfers UCT (barnetrygd) Norway Housing Benefit (bostøtte) 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers UCT Norway Social Economic assistance (økonomisk 1. Yes 2. Ad hoc 1. Prices 1. Annual HIC Cash transfers UCT stønad) Poland Family Allowance 1. Yes 2. Ad hoc 3. ad hoc 4. Other HIC Cash transfers UCT Portugal Social Integration Income (Rendimento Social 1. Yes 2. Ad hoc HIC Cash transfers UCT de Inserçã; RSI) Romania Guaranteed minimum income (Schema privind 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers UCT venitul minim garantat) Russia Federal social pension supplement* 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Social pensions Old age social (non-contributory) pensions Russia Monthly cash payment (federal)* 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Cash transfers UCT Russia Regional social pension supplement * 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Social pensions Old age social (non-contributory) pensions Russia Social pension provided in the framework of 1. Yes 1. Automatic 1. Prices UMIC Social pensions Old age social the state pension security (non-contributory) (non-contributory) pensions including old age* Russia Social pension provided in the framework of 1. Yes 1. Automatic 5. Prices and 5. Prices and UMIC Social pensions Disability pensions the state pension security (non-contributory) Wages Wages (non-contributory) including disability social pensions* Russia Social pension provided in the framework of 1. Yes 1. Automatic 5. Prices and 5. Prices and UMIC Social pensions Survivors pensions the state pension security (non-contributory) Wages Wages (non-contributory) including survivorship social pensions* Serbia Financing Social Assistance [formally known 1. Yes 1. Automatic 1. Prices 2. Semi-annual UMIC Cash transfers UCT as Material Support for Low-income House- holds Program (MOP)] Slovakia Minimum Income Scheme “Assistance in 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers UCT material needs” (Pomoc v hmotnej núdzi). Slovenia Financial social assistance (denarna socialna 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers UCT pomoč) Spain Non-contributory retirement pension (pensión 1. Yes 2. Ad hoc 5. Discretionary HIC Social pensions Old age social no contributiva de jubilación) (non-contributory) pensions Sweden Ekonomiskt Bistånd/Försörjningsstöd (mini- 1. Yes 2. Ad hoc 1. Prices 1. Annual HIC Cash transfers UCT mum income benefit) Switzerland Welfare program 1. Yes 2. Ad hoc 5. Prices and 4. Other HIC Cash transfers UCT Wages Tajikistan Targeted Social Assistance* 1. Yes 1. Automatic 1. Prices LMIC Cash transfers UCT 79 >> KEEP THE PACE Turkey CCT Education 1. Yes 2. Ad hoc 5. Discretionary UMIC Cash transfers CCT Ukraine Social assistance for low income families 1. Yes 2. Ad hoc LMIC Cash transfers CCT (Соціальна допомога малозабезпеченим сім’ям) United Kingdom Universal Credit 1. Yes 2. Ad hoc 1. Prices 1. Annual HIC Cash transfers UCT LAC LAC LAC LAC LAC LAC LAC LAC LAC Antigua Old-age Assitance Programme 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary HIC Social pensions Old age social and Barbuda (non-contributory) pensions; Disability pensions Antigua People’s Benefit Program 2. No HIC Cash transfers UCT and Barbuda Argentina Familias por la Inclusión Social 1. Yes 2. Ad hoc 6. Wages UMIC Cash transfers CCT and Macro- economic variables Argentina Non-contributory pension programme 1. Yes 1. Automatic 6. Wages 3. Quarterly UMIC Social pensions Old age social (Programa de pensiones no contributivas) and Macro- (non-contributory) pensions; Disability economic pensions variables Argentina Universal Pension for the Elderly (Pensión 1. Yes 1. Automatic 6. Wages 3. Quarterly UMIC Social pensions Old age social Universal para el Adulto Mayor (PUAM)) and Macro- (non-contributory) pensions economic variables Argentina Universal Child Allowance for Social Protection 1. Yes 1. Automatic 1. Prices 3. Quarterly UMIC Cash transfers CCT Argentina Programa de Ciudadanía Porteña 1. Yes 1. Automatic 2. Semi-annual UMIC Cash transfers CCT (Citizenship Program) Bahamas Invalidity Assistance 1. Yes 2. Ad hoc 5. Discretionary HIC Social pensions Disability pensions (non-contributory) Bahamas Old-age Non-contributory Pension 1. Yes 2. Ad hoc 5. Discretionary HIC Social pensions Old age social (non-contributory) pensions Barbados Old-age Asisstance Pension 1. Yes 1. Automatic 5. Prices and 1. Annual HIC Social pensions Old age social Wages (non-contributory) pensions; Disability pensions Belize Non Contributory Pension Program 2. No UMIC Social pensions Old age social (non-contributory) pensions; Disability pensions Bermuda Non-Contributory Pension 1. Yes 2. Ad hoc 1. Prices 5. Discretionary HIC Social pensions Old age social (non-contributory) pensions; Disability pensions Bolivia Juancito Pinto Grant 2. No LMIC Cash transfers CCT Bolivia Juana Azurduy de Padilla Mother-and-Child 2. No LMIC Cash transfers CCT Grant Bolivia Bonosol “Bono Solidario” 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary LMIC Social pensions Old age social (non-contributory) pensions Bolivia Renta Universal de Vejez “Renta Dignidad” 1. Yes 2. Ad hoc 4. Other LMIC Social pensions Old age social (non-contributory) pensions Brazil Bolsa Familia 1. Yes 2. Ad hoc 5. Discretionary UMIC Cash transfers CCT Brazil Cartão Alimentação (food card) 2. No UMIC Cash transfers CCT Brazil Continuous Benefit Programme or Benefício 1. Yes 1. Automatic 2. Wages 1. Annual UMIC Social pensions Old age social de Prestação Continuada da Assistência (non-contributory) pensions; Disability Social (BPC) pensions Brazil Previdencia Rural 1. Yes 1. Automatic 2. Wages 6. Monthly UMIC Social pensions Old age social (non-contributory) pensions; Disability pensions Brazil Programa Bolsa Verde 2. No UMIC Cash transfers CCT Chile Solidarity Chile 1. Yes 2. Ad hoc 1. Prices 1. Annual HIC Cash transfers CCT Chile Securities and Opportunities 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers CCT (Ethical Family Income) Chile Single Family Allowance 1. Yes 2. Ad hoc 5. Discretionary HIC Cash transfers CCT Colombia Families in Action 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Cash transfers CCT Colombia Conditional Subsidies for School Attendance 1. Yes 2. Ad hoc 5. Discretionary UMIC Cash transfers CCT Colombia Colombia Elderly Programme 1. Yes 2. Ad hoc UMIC Social pensions Old age social (non-contributory) pensions Costa Rica Avancemos 1. Yes 2. Ad hoc 5. Discretionary UMIC Cash transfers CCT Costa Rica Non-contributory pension scheme 1. Yes 2. Ad hoc UMIC Social pensions Old age social by basic amount* (non-contributory) pensions Dominica Public Assistance Programme 2. No UMIC Cash transfers CCT Dominican Republic Solidarity cash tranfer 1. Yes 2. Ad hoc UMIC Cash transfers CCT (PROGRESANDO CON SOLIDARIDA) Dominican Republic Improve yourself (ex Progressing 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary UMIC Cash transfers CCT with Solidarity) Ecuador Human Development Grant 1. Yes 2. Ad hoc UMIC Cash transfers CCT El Salvador Solidarity in Rural Communities 2. No LMIC Cash transfers CCT (formerly the Solidarity Network) Grenada Safety Net Advancement Project 3. Not available UMIC Cash transfers CCT Guatemala Mi Familia Progresa 2. No UMIC Cash transfers CCT Guatemala Social Allowance 1. Yes 2. Ad hoc UMIC Cash transfers CCT Guatemala Social Basket Food Package 2. No UMIC Cash transfers CCT 80 >> KEEP THE PACE Guyana Old Age Pension 1. Yes 2. Ad hoc 5. Discretionary UMIC Social pensions Old age social (non-contributory) pensions Haiti Ti Manman Cheri 1. Yes 2. Ad hoc LMIC Cash transfers CCT Honduras Family Allowance Programme (PRAF) 1. Yes 2. Ad hoc LMIC Cash transfers CCT Honduras Better Life Grant (Bono Vida Mejor) 1. Yes 2. Ad hoc LMIC Cash transfers CCT Honduras PRAF/IDB Tranche 3 1. Yes 2. Ad hoc LMIC Cash transfers CCT Jamaica Programme of Advancement through Health 1. Yes 2. Ad hoc 7. Prices 2. Semi-annual UMIC Cash transfers CCT and Education (PATH)46 and Macro- economic variables Mexico Prospera (formely known as Oportunidades 2. No UMIC Cash transfers CCT and before that as Progresa)47 Mexico Benito Juarez Scholarship for Well-being48 2. No 2. Ad hoc 1. Prices UMIC Cash transfers CCT Mexico Pension for Older People49 1. Yes 1. Automatic 1. Prices 1. Annual UMIC Social pensions Old age social (non-contributory) pensions Mexico Pension for the Well-Being of Older People 1. Yes 2. Ad hoc 1. Annual UMIC Social pensions Old age social (non-contributory) pensions Mexico Programme of Food Support for Adults 1. Yes UMIC Social pensions Old age social over 68 years old living in Mexico City (non-contributory) pensions Mexico Stimulus Program for the Universal 2. No 2. Ad hoc UMIC Cash transfers CCT Baccalaureate50 Nicaragua Sistema de Atención a Crisis 1. Yes 2. Ad hoc LMIC Cash transfers CCT (Crisis Response System) (2005-2006) Nicaragua Social Protection Network 1. Yes 2. Ad hoc LMIC Cash transfers CCT (Red de Protección Social; RPS)* Panama Opportunities Network 1. Yes 2. Ad hoc HIC Cash transfers CCT Paraguay Tekoporâ 1. Yes 5. Discretionary UMIC Cash transfers CCT Paraguay Abrazo 3. Not available UMIC Cash transfers CCT Peru Juntos 1. Yes 2. Ad hoc UMIC Cash transfers CCT Peru National Solidarity Assistance Programme 2. No UMIC Social pensions Old age social “Pension 65“ (2011-) (non-contributory) pensions St. Lucia Short Term Employment Programme 3. Not available UMIC Public works Cash for work Trinidad and obago Senior Citizens’Pension (ex Old Age Pension) 1. Yes 2. Ad hoc HIC Social pensions Old age social (2001-) (non-contributory) pensions Uruguay Non contributory pensions for older people and 1. Yes 1. Automatic 2. Wages 1. Annual HIC Social pensions Old age social the disabled (non-contributory) pensions; Disability pensions Uruguay Asignaciones Familiares - Plan Equidad (Family 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers CCT allowances - Equity Plan) Venezuela Great Mission in Elder Love 1. Yes 2. Ad hoc 2. Wages UMIC Social pensions Old age social (non-contributory) pensions MENA MENA MENA MENA MENA MENA MENA MENA MENA Algeria Special Allowance for School Children 1. Yes 2. Ad hoc 5. Discretionary LMIC Cash transfers CCT Djibouti National Program of Family Solidarity 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary LMIC Cash transfers CCT Egypt Takaful 1. Yes 2. Ad hoc LMIC Cash transfers CCT Iraq Social Protection Network 1. Yes 2. Ad hoc UMIC Cash transfers UCT Israel Supplementary Income for the Elderly 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers UCT (Income Support for the Elderly) Israel Child Allowance for Lone Parents 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers UCT Jordan World Food Programme E-card 1. Yes 2. Ad hoc 1. Prices UMIC Cash transfers UCT for purchase of food items Lebanon National Poverty Targeting Programme (NPTP) 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary LMIC Cash transfers UCT Libya Basic Pension Benefit 1. Yes 2. Ad hoc 2. Wages UMIC Social pensions Old age social (non-contributory) pensions; Disability pensions Morocco Tayssir Program (Cash Transfer 2. No LMIC Cash transfers CCT Programme for Children) Saudi Arabia Regular Assistance: Divorced, Widowed HIC Cash transfers UCT Women (Kanaf Financing)* Tunisia PNAFN Education Allowance Program 1. Yes 2. Ad hoc 3. ad hoc 1. Annual LMIC Cash transfers UCT West Bank and Cash Transfer Program* 2. No LMIC Cash transfers UCT Gaza NORTH AMERICA NORTH AMERICA NORTH AMERICA NORTH AMERICA N. AMERICA N. AMERICA N. A NORTH AMERICA NORTH AMERICA Canada Old Age Security (OAS) 1. Yes 1. Automatic 1. Prices 3. Quarterly HIC Social pensions Old age social Canada (non-contributory) pensions United States Canada Child Benefit (CCB) 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers CCT United States Supplemental Security Income (SSI) 1. Yes 1. Automatic 1. Prices 1. Annual HIC Cash transfers UCT Temporary Assistance for Needy 1. Yes 2. Ad hoc HIC Cash transfers UCT Families (TANF) SAR SAR SAR SAR SAR SAR SAR SAR SAR Afghanistan Martyrs and Disabled Pension Programme 1. Yes 2. Ad hoc 5. Discretionary LIC Social pensions Disability pensions (non-contributory) Bangladesh Old Age Allowance 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary LMIC Social pensions Old age social (non-contributory) pensions Bangladesh Stipend for primary students 1. Yes 2. Ad hoc LMIC Cash transfers CCT India Mahatma Gandhi National Rural 1. Yes 1. Automatic 1. Prices 1. Annual LMIC Public works Cash for work Employment Guarantee 81 >> KEEP THE PACE Maldives Old Age Basic Pensions Scheme 1. Yes 2. Ad hoc 1. Prices UMIC Social pensions Old age social (non-contributory) pensions Nepal Social Security allowance (Old Age Allowance) 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary LMIC Social pensions Old age social (non-contributory) pensions Nepal Social Security Allowance (child allowance) 1. Yes 2. Ad hoc 3. ad hoc 5. Discretionary LMIC Cash transfers UCT Pakistan Benazir Income Support Programme 1. Yes 2. Ad hoc 1. Prices 1. Annual LMIC Cash transfers UCT Sri Lanka Samrudhi 1. Yes 2. Ad hoc LMIC Cash transfers UCT Sri Lanka Public Assistance Monthly Allowance or Public 1. Yes 2. Ad hoc LMIC Cash transfers UCT Welfare Assistance Allowance (PAMA/Pin Padi) 1 —The benchmark indicator for China’s Dibao program could also fall under the “Prices and Macro-economic variables” category, if expenditures is added to price-based indexation. Additional research on current indexation practice should be further investigated. Note: *= key features on program benefit adjustments were limited and the team has used its best judgment based on the available information 82 >> KEEP THE PACE 83