Vietnam's Labor Market Institutions, Regulations, and Interventions: Helping People Grasp Work Opportunities in a Risky World

Over the past 30 years, Vietnam has experienced significant shifts of employment away from agriculture toward manufacturing and services, away from household enterprises toward registered and regulated businesses, and away from state-owned enterprises toward private firms. This paper argues that for these processes to continue in the future, appropriately designed and implemented labor market policies need to be in place, including labor market regulations that protect workers but do not inhibit creative destruction and creation of formal sector jobs; labor market interventions that improve workers' human capital, eliminate information asymmetries, and are fiscally sustainable; and labor market institutions that give voice to workers and employers. As a part of all of these measures, Vietnam will also have to renew its efforts to integrate vulnerable groups into the labor market.


Introduction
Vietnam's current labor market policies have evolved in the context of its long and ongoing transition from a mainly planned to a modern, market-led economy with a fast-growing middle class. This transition started in the mid-1980s with the doi moi ("renovation") reforms. Over the last 30 years, it has led to significant shifts of employment away from agriculture toward manufacturing and services; away from household enterprises toward registered and regulated businesses; and away from state owned enterprises (SOEs) toward private domestic and foreign-owned firms (cf. McCaig and Pavcnik, 2013).
So far, Vietnam has been remarkably successful in managing its transition. Observed through an employment lens, regional disparities persist but unemployment has been low, wage growth pronounced and -at least among private sector workers -wage inequality has been declining (cf. Pierre, 2012, Imbert, 2013, and Sakellariou and Fang, 2014. Moreover, there are few signs that Vietnam's labor market policies have thus far constrained the pace of the country's growth. In fact, results from World Bank investment climate surveys show that a smaller proportion of firms identifies labor market regulations as a constraint to doing business in Vietnam than in Cambodia, Malaysia, Thailand and many other countries in the East Asia and Pacific (EAP) region (cf. World Bank, 2014a).
At the same time, Vietnam's transition from central planning to a market economy is far from complete. In many sectors SOEs remain dominant and workers with access to state jobs continue to enjoy rents. In 2008, wages were 12 to 14 percent higher in the public than in the private sector even for workers with comparable observable characteristics (cf. Phan and Coxhead, 2013). Moreover, Vietnam's transition from an agrarian, rural and informal economy, to an urban, manufacturing and services-based and formal economy is not complete either (cf. World Bank, 2014b). According to an investment climate survey, in Vietnam more enterprises see the practices of competitors in the informal economy as the biggest obstacle for doing business than is the case for any other element of the business environment apart from access to finance (cf. World Bank, 2009). Moreover most working people in Vietnam remain engaged in agrarian, rural and informal work. Table 1 breaks Vietnam's workforce down by sector and type of employment. Almost half of the country's workforce is still engaged in small-scale agriculture. Another third of the workforce is active in non-agricultural household enterprises. With regard to types of employment, Table 1 shows that a majority of working people actually work outside of the employer-employee, "dependent employment" model on which most labor market policies are based: 45 percent of workers are self-employed and 17 percent are unpaid family workers. In contrast, only 35 percent of working people are wage and salaried "dependent" workers. Among these, almost half do not hold a written contract of employment. As Vietnam's transition continues and market forces play an increasingly dominant role, labor markets can become more important places of economic opportunities for working people. But these new opportunities will be accompanied by new risks as well as by new expectations from citizens about how their government can help to mitigate risks. As a result, managing and mitigating risks will become an ever more important element of the social contract. Combined with other forms of social protection, appropriately designed and implemented labor market policies -labor market regulations, interventions, and institutions -have the potential to help people better grasp and take advantage of economic opportunities in a risky world. More specifically, labor market policies have the potential to help lower an individual's probability of experiencing adverse events such as job displacement that might result in earnings and consequently welfare losses. When shocks occur, policies can also help reduce the size of earnings losses, provide voice to workers and other labor market actors and strengthen growth by improving access to productive opportunities, as well as by facilitating the movement of labor and human capital to where they are most efficiently deployed. Finally, they can contribute to ensuring equality of opportunity on the labor market and basic protection for those vulnerable to exclusion.
On the downside, labor market policies that are not appropriately designed and implemented -i.e.
policies that are insufficient, excessive or poorly implemented -may increase both individual-level labor market risks (like job displacements) and labor market risk affecting society as a whole (such as strikes, lock-outs and labor unrest). Moreover, such policies may slow down or bring to a halt the movement of labor and human capital across geography, sectors, and types of work. For a lower middle income country like Vietnam, labor and human capital are two of the most crucial factors of production and a slow-down in the movement of labor and human capital will slow down overall development, raising the risk of the country getting stuck in a middle income trap (cf. Gill and Kharas, 2007).
Today, labor market polices do no not yet directly affect a majority of workers in Vietnam because the majority are outside the regulated labor market. However, rather than cause for complacency, the low effective coverage of policies should motivate decision makers to take action. First of all, even today labor market polices might play a role that goes beyond their immediate coverage. These policies influence firms' decision to offer jobs inside or outside the regulated labor market and workers' choices to work in either segment of the labor market. Moreover, labor market polices can have indirect effects even on workers outside the regulated labor market. An example of this is the so called "lighthouse effect" of statutory minimum wages: if sector wages inside the regulated labor market serve as the economy-wide reference point, changes in the minimum wage will influence wages and employment even outside the regulated labor market. What is more, the relatively limited direct coverage of labor policies coupled with Vietnam's ongoing but incomplete transition presents a window of opportunity to discuss the potential effects of different labor market policies on the country's development. If this discussion does not take place now and forward-thinking labor market policies are not put into place soon, as more labor moves to factories and firms and these continue to grow, weaknesses that are already apparent today will evolve into even bigger problems that will be much harder to fix tomorrow.
The rest of this paper is structured as follows: Section 2 provides findings on the current situation of Vietnam's labor market institutions, regulations and interventions, as well as the extent of opportunity and basic protections for those vulnerable to exclusion. Section 3 deliberates where Vietnam wants to go with respect to its labor market. Finally, Section 4 concludes by discussing what labor market policies will be helpful for Vietnam to achieve its goals.

Labor Market Regulation
Labor market regulations like minimum wages and employment protection legislation (EPL) set the legal parameters of employment and imply costs to employers and workers. Minimum wages, for instance, are widely used to correct labor market imperfections that can result in an unfair distribution of gains in productivity. However, when set incorrectly, minimum wage regulation can result in the loss of jobs and decreased firm competitiveness.
Vietnam first set a statutory minimum wage in 1992 that exclusively applied to foreign companies.
Over time, it also established a minimum wage for domestic firms. The country's current minimum These are often seen as a tool for assuring living standards. In fact, it has been the Government of Vietnam's stated ambition to raise minimum wages to a level that allows households to reach the "minimum monthly living standard" in 2018. The consumption necessary to obtain the minimum monthly living standard was measured in a study by the Institute for Workers and Trade Unions (2014) which is part of the Vietnam General Confederation of Labour (VGCL). As detailed in the study, current minimum wage levels outside the general government sector fall far short of the minimum monthly living standard. However, this does not imply that households with minimum wage earners necessarily fall below the poverty line. Figure 1 shows the ratios of labor earnings to the poverty line -as defined by the General Statistics Office (GSO) and the World Bank (cf. Demombynes and Vu, 2015) -for different household compositions that vary according to two dimensions. First, the number of household members varies from one to six. Second, a varying number of household members holds a minimum wage job (outside the general government sector) while the remaining household members are assumed to be dependents and not part of the labor force. Figure 1 makes it clear that while households with one minimum wage earner and at least three additional dependents will be below the poverty line, households with at least two minimum wage earners or households with one minimum wage earner and less than three additional household members who consume all their labor earnings, are able to afford a level of consumption above the poverty line. The group of households with minimum labor earnings above the poverty line includes what can be considered Vietnam's average or representative household: a household with two working people and two dependents. 1 While Figure 1 focuses on region 1, patterns look broadly similar for the three other minimum wage regions.
In any case, policies and regulatory instruments other than the minimum wage are usually seen as more appropriate for assuring that people who work full time are able to live above (or even substantially above) the poverty line. With regard to the appropriate role of the minimum wage, moderate minimum wages have been shown to be effective in correcting labor market failures and have thereby led to an increase in labor earnings and employment. To assess whether Vietnam's current minimum wage levels are insufficient, moderate or relatively excessive, Figure 2 depicts the ratios of minimum to average and median wages in Vietnam and selected countries in the OECD and the EAP region. As Vietnam is an outlier in that it has distinct minimum wage levels for the general government sector and the rest of the labor market, its minimum wage "bites" are depicted in this figure separately for the general government and other sectors. Dube (2014) proposes as a rule of thumb that the minimum wage should be set at about half of the median wage and that there are at least three reasons to pay attention to measures of the minimum wage "bite". First, a comparison of the minimum wage to the average or especially the median wage offers a guide for how binding a particular minimum wage level is likely to be. Second, the minimum wage "bite" also provides a benchmark for determining how high or low a minimum wage is across time periods or countries. Third, it gives a natural reference point for judging whether the level of the  Reading aid: A household with three members ("3" on the horizontal axis) one of whom is employed at the minimum wage while the two others are not employed ("1" on the vertical axis") has labor earnings that are higher than the poverty line for a three person household (indicated by blue bar); for a household with four members ("4" on the horizontal axis) one of whom is employed at the minimum wage while the three others are not employed ("1" on the vertical axis") labor earnings are lower than the poverty line for a four person household (indicated by orange bar).
As is clear from Figure 2, Vietnam's general government sector minimum wages are low. For the remainder of the labor market, however, the picture is different. Across the four regions, the average ratio of the statutory minimum wage to the mean wage is 50 percent for workers outside the general government sector. The average ratio of the minimum to the median wage outside this sector is 58 percent. Both numbers are among the highest statutory minimum wage "bites" found anywhere in the world and are only exceeded by a small number of countries in either the OECD or the EAP region.   Minimum relative to median wages assumption that Vietnam will fulfill its stated ambition to raise minimum wages to the level of the minimum monthly living standard in 2018. By way of comparison, the growth of real labor productivity in the state, non-state domestic and FDI sectors are also depicted. Finally, Figure   As is evident from Figure 3, Vietnam's non-government domestic sector minimum wages have been rising rapidly. Their real rate of increase has been in or near the double digits for each of the last ten years. Growth was particularly rapid in 2012 when minimum wages in the domestic sector and the FDI sector were aligned. The prolonged period of high growth rates has led to today's situation where minimum wage levels are high by international standards (and therefore a threat for job creation and  firm competitiveness). In spite of this, if current plans are put into practice, the growth of private sector minimum wages will continue to accelerate at least until 2018. At the same time, productivity growth has lagged behind. Unless productivity growth picks up markedly over the coming years, it will further decouple from the pace of growth in statutory minimum wages. 3

Figure 4 -Index for strictness of EPL legislation
Sources: World Bank staff calculations partly based on World Bank (2014a) and OECD data.
Notes: Data are for 2010; higher scores represent stricter employment protection legislation.
As mentioned above, employment protection legislation is another category of labor market regulations. It consists mainly of mandated risk-management regulations that lower the probability of The OECD calculates an EPL index that compares some of the most critical of EPL components and the costs they impart on employers: first, protections of permanent workers against individual dismissal; second, additional requirements for collective dismissal; and third, regulations and restrictions on temporary forms of employment. The index ranges from zero to six, with zero being the least restrictive and six the most restrictive legislation. The values for Vietnam and selected countries in the OECD and the EAP region are shown in Figure  The most dramatic changes occurred with respect to fixed term contract regulations. The index became more stringent as a consequence of new regulations for outsourcing contracts or labor leasing enterprises. In the past, there were no specific regulations for labor leasing enterprises but regulations regarding fixed-term contracts applied. Under the new regulation, labor leasing enterprises have more restrictions, and outsourcing contracts are limited to a maximum cumulative duration of 12 months (and not 72 months as is the case for fixed-term contracts). As is evident from Figure 4, these reforms meant that Vietnam moved from being a country with moderately stringent restrictions on temporary forms of employment to one that is very restrictive in regulating this type of employment. Overly tight regulation of temporary forms of employment can hinder the efficient allocation of workers. A company that is struggling to shift its business model in times of structural change needs flexibility in reorganizing its labor inputs. Similarly, as large SOEs make way for nimbler sorts of production, new job opportunities can be created. But many of these jobs are more likely to be created by labor leasing -12 -agencies than by other big and slow-acting enterprises. If labor leasing agencies are severely restricted in their operations, these kinds of jobs might not appear at all. Figure 4 shows that Vietnam is among the group of countries with the most restrictive employment protection regulation. As overly stringent EPL implies considerable costs for both employers and workers, this constitutes a considerable obstacle to the process of "creative destruction", and the associated movement of labor that any modern market economy needs in order to thrive. While it is difficult to quantify the effects of overly restrictive employment protection regulations beyond the synthetic representation in Figure 4, a growing body of evidence from the EAP region and beyond shows that they are often associated with lower employment, especially of young workers, women, new entrants to the labor market, and unskilled workers (cf. Heckman and Pages 2004;OECD 2004;Perry et al. 2007;Packard, Koettl, and Montenegro 2012;World Bank, 2014a). These same groups tend to be among the first to be laid off when labor costs rise or to be blocked from entry altogether (cf. Lustig and McLeod 1996). Moreover, Vietnam's strict EPL regime might be one of the main reasons why so many workers outside of the agriculture sector and in urban areas remain outside the regulated labor market: evidence from countries around the world demonstrates that more restrictive EPL raises the share of self-employment and decreases dependent employment in low-and middleincome countries (cf. Addison and Teixeira, 2001;Betcherman, Luinstra, and Ogawa, 2001;Haltiwanger, Scarpetta, and Vodopivec, 2003).

Labor Market Interventions
Labor market interventions are deployed by the state to make up for market shortcomings, such as uneven distribution of information that impedes the best matches between employers and workers, or the inability of private financial markets to efficiently insure losses from the risk of unemployment.
The resulting costs are borne by taxpayers irrespective of whether they participate in the labor market.
Interventions can include both "active" labor market programs like training and job search assistance and "passive" unemployment benefits and other forms of social insurance.
The objective of passive labor market programs is to protect individuals against the risks to their consumption from job loss for a period of job search. This is basically the same objective as that of severance pay arrangements and other elements of EPL, creating a potential substitutability of employer provided severance and unemployment benefits and other forms of social insurance.
Generally speaking, however, unemployment insurance schemes are more efficient and secure than severance pay arrangements in pooling risks related to job loss. This is because their risk pools extend beyond the boundaries of a single enterprise. With regard to the effects of passive labor market programs on unemployment and other labor market outcomes, these are in principle ambiguous. By their intended purpose, passive programs can remove the urgency of finding new work and improve the quality of employer-employee matches. However, if unemployment benefits are overly generous or poorly designed, they can perversely lower peoples' motivation to look for and accept employment.

Qualifying conditions
Thailand 50 percent of the average daily wage in the highest paid 3 months during the 9 months before unemployment; the maximum daily benefit is B 250 6 months in any 1 year 6-month contribution within 15 months before unemployment

China
Flat benefit level set by the local authorities; higher than the local public assistance benefit, but lower than the local minimum wage

12-24 months, depending on applicant's years of contribution
Maximum of 1 year with less than 5 years of contributions; maximum of 1.5 years with 5 or more but less than 10 years of contributions; maximum of 2 years with 10 or more years of contributions Vietnam 60 percent of the average monthly earnings in the 6 months before unemployment 3-12 months, varying by contribution 3 months with 12-35 months of contributions; 6 months with 36-71 months of contributions; 9 months with 72-143 months of contributions; 12 months with 144 months of contributions or more

Source: World Bank (2014a).
Together with Thailand and China, Vietnam is one of only three developing countries in the EAP region with a national unemployment insurance scheme. Vietnam's unemployment insurance scheme was introduced in 2009. Its main characteristics as well as those of the Thai and Chinese schemes are summarized in Table 2. As is evident from the table, unemployment insurance benefits in Vietnam are relatively generous. At 60 percent of the average monthly earnings in the six months before unemployment, the benefit level or replacement rate (the portion of previous income replaced by insurance benefits) is higher than those of Thailand or China. As is common in many countries, in Vietnam the duration entitlement of unemployment benefits rises in a step-wise fashion with the worker's years of contributions. It ranges from three months for a worker who contributed for 12-36 months to six months for a worker with 36-71 months of contributions to nine months with 72-143 months of contributions. A worker who contributed for 144 months or more can draw unemployment benefits for 12 months.
In Latin America, Eastern Europe and Central Asia, labor-intensive public works and other active labor market programs -targeted training programs, wage subsidies, and employment search assistance -have long been a mainstay of social protection and labor policy. In contrast, in low and middle-income countries in the EAP region, they are relatively rare. But even in this region Vietnam is somewhat of an outlier as it has very few active labor market policies. Moreover, those policies that do exist -like public employment programs or programs to support employment changes for laborers in rural areas as described in the Employment Law -often have low coverage and funding. The lack of more comprehensive active labor market polices might impede the best matches between employers and workers and might even create issues that go beyond labor market policies but touch upon social protection more broadly.
Most labor market interventions are financed mainly by mandatory contributions from employers and workers. As a result, mandatory contributions paid by formal workers and employers tend to be substantial. In case workers and/or employers assign little value to interventions financed by these contributions, the mandate to contribute creates a "tax wedge" between the cost of labor to a firm and the take-home pay of employees. The size of this wedge can influence both the demand for and supply of labor and human capital. The impact on incentives is greater where there is a mismatch between the actual labor market interventions and what is desired by workers, where government has a poor track record of performance and where few workers believe they will ever benefit from the interventions. If the perceived tax wedge is large and net wages are too low, workers may be discouraged from participating in the regulated part of the labor market, finding outside optionsunemployment or irregular, informal work -more attractive. Similarly, because taxes increase the costs of employing someone, they can discourage employers from hiring, or offering legislated protections through a written employment contract.  Figure 5 depicts the labor tax wedge in Vietnam and selected countries in the OECD and the EAP region. The figure shows that, at slightly below 30 percent, Vietnam's labor tax wedge is lower than for instance's China's, similar to the levels observed for Australia and Ireland and higher than those found in the Philippines and Malaysia. Overall, Vietnam is placed around the mid-point of the regional distribution of the labor tax wedge. However, its wedge is relatively high when compared to those of other countries with a similar level of economic development. Moreover, as Vietnam has few active labor market policies, and the country's unemployment insurance system has only been in place for a few years, workers might not assign much value to the interventions financed through the labor tax or might doubt whether benefits will be available when they are needed. Thus one might conjecture that they are more likely to consider mandatory contributions as a tax rather than as part of their incentives. As a result, it seems unfeasible to finance growing demands for social protection measures through still higher contributions from employers and workers alone. Instead, more reliance on general revenues to fund coverage expansion is likely to result in better and more equitable outcomes.

Labor Market Institutions
Institutions are the legislated structures, recognized norms, and agreed procedures by which interested parties exert their influence and make and carry out decisions that shape the regulations and interventions covered in earlier sections. In the labor market, the best example of an institution is the space afforded in the legal code of many countries for industrial relations, collective bargaining, labor unions and employer associations. Second, unions act as a transmission mechanism between the CPV and working people and are supposed to represent these people in the highest levels of government decision making. Third, unions are tasked with protecting workers' rights and representing workers' interests vis-à-vis firm management and providing welfare services to workers. The explicit roles for employers' associations are similarly varied.   Figure 7 also shows that almost no strikes occur in SOEs. In the SOE sector, wages and benefits are relatively high, and relations between workers' representatives and management are close. In contrast, the foreign-owned sector appears particularly exposed to strikes. In addition to strikes, the sector has also been plagued by high labor turnover and absenteeism (cf. Chi and van den Broek, 2013 While the high and rising number of wildcat strikes are a symptom of the weaknesses of Vietnam's industrial relations system, their consequences reach much farther. Strikes by workers (and lock-outs by employers) are typically taken as "last resort" measures, when no agreements can be reached on pay, working conditions or equally fundamental issues. Strikes are in most contexts an outcome of negotiations failing. Thus the pattern shown in Figure 7 raises the concern that the institutions for  be a lack of functioning institutions for collective bargaining; those institutions that exist lack credibility in the eyes of workers as well as employers.

Equality of Opportunity and Basic Protection for Those Vulnerable to Exclusion
Appropriately designed and implemented labor market policies can not only help people grasp economic opportunities in a risky world, but also contribute to ensuring equality of opportunity on the labor market and basic protection for those vulnerable to exclusion. A number of different groups are potentially vulnerable to exclusion on the labor market. These include ethnic minorities, displaced persons and migrants, and the disabled. The largest group that is potentially vulnerable to exclusion on the labor market are women. Put differently, the successful integration of women into the labor market might promise the greatest payoffs in terms of increased productivity and improved social cohesion. Some commentators have even argued that at least one factor contributing to the low ratio of female to male births in many Asian countries is that rational and forward-looking parents see greater labor market opportunities for men and therefore prefer male offspring (cf. Qian, 2008).  Figure 8) is the ratio of the female to the male labor force participation rate and the second (shown in Figure 9) the average wage gap between women and men.
Over the last 15 years Vietnam's female labor force participation rate has consistently been lower than the corresponding figure for men. This result is not surprising as male labor force participation rates tend to be comparatively higher almost throughout the world. In fact, with a gap that is stable and at around 10 percent relatively modest, Vietnam has achieved greater gender parity in terms of labor force participation than many other countries. Among the countries from the EAP region whose time series are depicted in Figure 8, only Cambodia, Myanmar and the Lao People's Democratic Republic have ratios of female to male labor force participation rates that exceed Vietnam's.   While internationally comparable and recent data on the gender wage gap is hard to obtain, Figure 9 suggests that Vietnam's ratio of female to male nominal wages is not far from the regional average.
Whereas almost no gender wage gap is recorded for Thailand and the Philippines, Vietnam's gap is comparable to the wage gaps of Myanmar and Mongolia and smaller than that of Cambodia.
Overall, while no particularly pronounced disparities or trends in terms of gender differentials in the probability to participate in the labor force or wage rates are apparent, current policies appear not to lead to a convergence of outcomes between men and women. This is worrisome. But it also means that a complete integration of women into Vietnam's labor market promises substantial payoffs in terms of increased productivity and improved social cohesion.

Discussion
So far, Vietnam has been very successful in managing three secular processes. The first has been the country's long transition from a mainly planned to a modern, market-led economy with a larger middle class. Second, urbanization has led to the movement of people and jobs from rural villages to urban centers. Finally, through formalization relatively insecure and unproductive jobs have begun to be replaced by others that offer better protection and more productive opportunities.
But a lot remains to be done. In 2015, the majority of working people in Vietnam were still engaged outside of the employer-employee model on which most labor market policies are based. The typical worker in Vietnam has relatively little education, lives in a rural area and is active in family farming, as a self-employed worker or in a household enterprise. If Vietnam is successful in continuing and accelerating the change process it started in the mid-1980s with the doi moi reforms, a generation from now its structure of employment will look completely different from today: a continuing transition from planned to market economy will strengthen the private sector. Rapid urbanization will bring even more workers from farms and rural workshops to jobs in the manufacturing and service sectors in big cities. Increased formalization will create labor markets and will replace small-scale household enterprises with more productive and better-paying jobs in the private domestic and FDI sector.
Vietnam also has the potential to develop a more socially progressive and equal society with narrower gender gaps, equality of opportunity on the labor market and basic protection for those vulnerable to exclusion.
However, it is far from certain that the change process started in the mid-1980s with the doi moi reforms will actually continue or accelerate in the future and this vision of Vietnam a generation from now will become reality. One condition will be avoiding getting stuck in the middle income trap, by maintaining the dynamism and mobility of labor and human capital across geography, sectors, and types of work. This movement will only happen if appropriately designed and implemented labor market policies are in place. Such labor market policies would include labor market regulations that protect workers but do not inhibit creative destruction and creation of formal sector jobs; labor market interventions that improve workers' human capital, eliminate some of the information asymmetries that delay and frustrate "matching" in the labor market and are fiscally sustainable; and labor market institutions that give voice to both workers and employers and lead to equitable, balanced and productive outcomes. A cross-cutting objective would be for Vietnam to renew its efforts to better integrate vulnerable groups like women, the disabled and minorities into the labor market.

Conclusions
If the ongoing transition, urbanization and formalization processes are to continue and accelerate, a generation from now the typical worker in Vietnam will likely be better-educated, live in a city and be an employee of a private domestic or FDI enterprise. But will these processes continue and accelerate, and will the positive vision of work in Vietnam a generation from now become reality? This will in part depend on the labor market policies that Vietnam pursues. More specifically, it will depend on whether the country uses the current window of opportunity to address weaknesses that are already apparent now before they become an even bigger problem once more people transition from farms to factories and firms. As shown in Figure 10, Vietnam today is right at the threshold of a transition out of a mainly agrarian and to a non-agrarian economy.
In order to understand in more detail what labor market policies have the potential to be effective in providing social protection from risks to people's earnings from work and influencing the demand for and supply of labor and human capital in Vietnam, it is helpful to once again distinguish between labor market regulations, interventions, and institutions. With regard to labor market regulations, appropriately designed and implemented policies will help Vietnam's people better grasp economic opportunities in a risky world. The minimum wage, in particular, will be better able to fulfill this role if it is used as an instrument to correct market failures in distributing gains in productivity between employers and employees, rather than as an instrument to fight poverty or guarantee a certain level of living standard. Other policy and regulatory instruments are much more effective -possibly in combination with the minimum wage -for assuring that people who work full time are able to live above (or rather substantially above) the poverty line. In contrast, the global evidence on the effectiveness of minimum wages in reducing poverty and inequality is checkered at best. At the same time, moderate minimum wages have been effective in correcting labor market failures caused for instance by monopsonistic competition or information asymmetry. In the short run, Vietnam's private sector would benefit if the country moderated the growth of minimum wage adjustments. In the medium term, stakeholders should refocus their attention from monthly to hourly minimum wages. This will help shift the attention from a desire to use minimum wage adjustments as a tool to reduce poverty and guarantee minimum living standards to one for addressing efficiency issues in the labor market and improving outcomes for employers and workers. Existing plans to anchor minimum wage adjustments in real productivity growth instead of headline inflation rates would also be helpful and should be put into place.
Vietnam' EPL has reached a worrying level of stringency. The country could benefit from streamlining and clarifying existing rules and regulations. In addition to that, it could make better use of the potential substitutability of employer provided severance and unemployment insurance. Generally speaking, unemployment insurance schemes are more efficient and secure than severance pay arrangements in pooling risks related to job loss. This is because their risk pools extend beyond the boundaries of a single enterprise. The potential substitutability of severance and unemployment insurance is apparently already recognized in Articles 48 and 49 of the 2012 version of Vietnam's Labor Code (on severance and redundancy pay, respectively) but the possibilities it opens could be further exploited. Moreover, Vietnam would benefit from loosening the regulation of labor leasing enterprises and temporary and outsourcing contracts. Such reforms will give firms more options than mass worker displacements in case their labor inputs become unviable. They will also give workers that have been displaced more opportunities to find a new job at a labor leasing company. More generally, they are likely to help foster the mobility and efficient allocation of workers and human capital and make the regulated part of the labor market relatively more attractive for both firms and workers.
Concerning labor market interventions, Vietnam should build on its pioneering role among developing economies in the EAP region and continue to expand its unemployment insurance scheme. At the same time, labor market interventions tend to have important social protection and fiscal implications. Therefore, Vietnam should keep close track of the fiscal impact of any expansion of unemployment insurance. It should also monitor the labor tax wedge and explore options of financing social protection, health and other programs out of general tax revenue. Internationally, it is good practice to have the portion of benefits that is designed to keep people out of desperation or to redistribute wealth financed through general revenues. In contrast, any "income replacement" that goes above and beyond such measures should be actuarially related to actual contributions by employers and/or employees.
Even in the EAP region Vietnam remains somewhat of an outlier in that it makes only very limited use of active labor market policies. The international evidence on the effectiveness of active labor -25 -market policies in reducing labor-related risk and smoothing job-to-job transitions is mixed, partly because many of these programs are only able to reach people who have lost registered forms of work (cf. World Bank, 2012b). Still, Vietnam should explore opportunities for modestly and prudently increasing its commitment to active labor market policies. This might involve both the expansion of existing programs like public employment policies or policies to support employment changes for laborers in rural areas or the creation of new programs.
Vietnam should also renew its efforts to better integrate vulnerable groups into the labor market. In this context, it should strengthen its rules on equal employment opportunities in the Labor Code.
Besides, it should explore other options for making sure that women, minorities, the disabled and other vulnerable groups can participate productively in the labor market. If this is successfully achieved, it will unleash the productive potential in terms of labor and human capital of these groups and promote overall growth. It will also have broader benefits for society by preventing discrimination and achieving equal pay for equal work. Enhanced economic opportunities for women might even help normalize the sex ratio at birth. Second, the government of Vietnam was given a more active role in facilitating collective bargaining.
Third, a National Wage Council as a new tripartite institution centrally involved in the setting of minimum wages was created. More generally, it seems that reforms of labor market regulations covering minimum wages and EPL (including more stringent regulations of labor leasing enterprises and more far-reaching restrictions of outsourcing contracts) were used as substitutes to bring about outcomes that would be ideally the result of firm-or sector-level collective bargaining agreements.
Today, Vietnam's Labor Code appears to bear the weight of weak industrial relations institutions.
Rapidly rising minimum wages and more stringent EPL have been used to address insufficient power of workers to negotiate adequate wages and assure fair and decent treatment by employers. This has resulted in a "one size fits all" set of regulatory parameters that fail to respond nimbly to the varying interests and demands of workers and employers and are not necessarily adequate across all sectors or geographies or types of work. Moreover, legislated parameters are difficult to alter once in place and can potentially result in the segmentation of Vietnam's labor market between insiders and outsiders.
In many ways, is seems as if only the symptoms caused by the dysfunction of collective bargaining mechanisms and the industrial relations system more generally have been treated. A real cure will only be possible if instead the underlying causes are addressed.
In the short term, Vietnam should continue to strengthen efforts to increase the capacity of unions and employers' federations through awareness campaigns, training in collective bargaining, etc. In recent years, the VGCL has begun attempts to form enterprise unions independent from management, to increase the capacity of such grassroots unions and to assure that unions actually represent workers' interests. These efforts should be supported and strengthened. Moreover, Vietnam should reinforce the capacity of its system for resolving industrial relations disputes. The capacity of mediators and arbiters appears particularly weak. Many of these work part-time, are overburdened with multiple and sometimes conflicting tasks or have insufficient professional training in collective bargaining and dispute resolution.
In the medium term, Vietnam could consider more fundamental measures. The VGCL and its affiliated unions could bar company executives and managers from principal union positions (like being head of an enterprise union). This would follow the example of many other countries -for instance Singapore -where company executives and managers cannot obtain principal union positions or sometimes even union membership. Besides, Vietnam should consider further increasing the penalties for violations of the Labor Code and strengthening the enforcement of its provisions. There is anecdotal evidence that even though penalties have already been increased, they are still being priced in by some firms as a cost of doing business but do not lead to real behavioral changes (cf. Schweisshelm, 2014). Finally, Vietnam should continue to simplify and streamline rules and regulations for resolving industrial relations disputes including but not limited to strikes and lock-outs.
Institutional reforms that allow the creation of firm-level works councils or labor-management councils as found in many European countries and Korea could also be considered. Labormanagement councils that bring together employer and employee representatives might help facilitate information sharing, promote employees' participation in decision-making processes and build trust between employers and employees.
In the long term, Vietnam should consider measures that might necessitate fundamental institutional reforms but could improve the quality of labor market institutions in a more sustainable way. The most fundamental package of reforms might include the streamlining of Vietnam's industrial relations -27 -system. This might involve a separation of the competing and partly contradictory roles of trade unions -to represent the CPV in enterprises, to provide welfare to workers, and to safeguard workers' rights and interests. Instead, unions could become independent of the government and the CPV and first and foremost concentrate on representing the rights and interests of workers. The provision of welfare to workers might remain as a secondary objective.
Today, China's All-China Federation of Trade Unions (ACFTU) has a comparable role to Vietnam's VGCL and China has similarly experienced informal wildcat strikes and frequent government intervention as second-best solutions when industrial partners failed to solve issues of mutual interest.
Moreover, while wildcat strikes in Vietnam have overwhelmingly been peaceful, similar situations in China have often turned confrontational or even outright violent (cf. Lee, 2006). A better institutional role model for Vietnam might be Singapore. The Singapore National Trade Union Congress (SNTUC) is the only national federation of trade unions in Singapore and its objectives are clearly defined in terms of safeguarding workers' rights and interests in a non-adversarial way. Today, Singapore's workers enjoy a very high standard of living; after a period of turmoil in the 1960s the incidence of strikes has declined practically to zero. 4

Appendix A: Vietnam's Minimum Wages and Wage Distribution
Comparisons of the statutory minimum wage to the mean and median wage give some indication of the "bite" of the minimum wage. However, they do not relate the minimum wage to the whole wage distribution but only to two of its location parameters. To assess how important the minimum wage is to the wage distribution as a whole, it is useful to draw kernel density plots of the distribution and the cumulative distribution of actual wages and compare these to the statutory minimum wage. As explained in more detail in Kristensen and Cunningham (2006), kernel density plots of wages smooth a line between each wage observation along the distribution or group observations within a certain part of the wage distribution in order to obtain an estimated density. Cumulative kernel density plots in turn plot the integral of kernel densities. If kernel density plots show an extremely high concentration at a particular wage and/or cumulative density plots a jump at the same wage -a "spike" -this indicates that wages are not randomly distributed, but that something is affecting the wage distribution and causing a clustering at that particular wage level. If that wage level coincides with the statutory minimum wage, then there is a good chance that the minimum wage legislation is the factor behind the spike in the kernel density plot.
As again detailed in Kristensen and Cunningham (2006), Kernel density plots can test if a minimum wage is binding and give insights into whether it is enforced. First, a minimum wage is deemed binding if wages in an otherwise continuous distribution are "pulled up" to the level of the statutory minimum.
If a minimum is set below the market clearing wage, it may still be binding if it is used as a reference point for setting wages higher in the wage distribution. A minimum wage can bind wages both in the formal sector and for informal workers (through the so-called "lighthouse effect"). Second, a minimum wage is actually being enforced if it is binding and all individuals covered by the minimum wage legislation receive a salary equal to or higher than the minimum wage. If a minimum wage falls below the market clearing level, we cannot determine whether or not it is being enforced. Vietnam's minimum wage for four groups of workers: those of SOEs, foreign-owned firms and private domestic firms with and without an employment contract. To account for the regional variation in minimum wages, all monthly earnings are expressed relative to regional minimum wages, i.e., earnings exceeding one are higher than the minimum wage while those below one are below the minimum wage. Vertical lines depict multiples of the minimum wage. In all four sectors considered here, rounding errors in how people report earnings in the labor force survey make the analysis difficult and are responsible for some of the most notable spikes in the distribution and cumulative distribution of monthly earnings. Still, Figure A.1 suggests that the minimum wage might be binding at least among workers with an employment contract in the private domestic sector and in the FDI sector. In both cases, there is something like a spike in the distribution of monthly earnings around the level of the minimum wage. Also for both sectors, few workers earn wages below the minimum wage so enforcement also appears high. Wages in the SOE sector are generally much more elevated than what is prescribed by the minimum wage rules and so we cannot say whether minimum wages are being enforced in the sector. There is not much evidence for a lighthouse effect of minimum wages on workers in the private domestic sector without a contract. A relatively large portion of these workers report wages that are lower than the minimum wage and there is no obvious spike in their (cumulative) wage distribution at the minimum wage level. nor the right to strike. In the absence of those rights, the government needs to intervene frequently if it wants to achieve its social policy goals.
In the 1950s and 1960s, Singapore also experienced a period of labor turmoil. Strikes were frequent and at times turned violent. But over the last five decades the incidence of strikes in the city state has declined practically to zero. In parallel, the living standards of the country's workers have gradually The work of the SNTUC and the remainder of the Singaporean labor movement focuses on three elements: First, workplace representation and protection of workers including collective bargaining and grievance handling. Second, improvements to workers' quality of life through the provision of a range of services such as access to recreational and leisure facilities. And third, political advocacy on behalf of workers. The SNTUC's leadership of the labor movement has been marked by pragmatism.
For example, in 1969 the SNTUC agreed to abandon an adversarial approach to industrial relations so as to make Singapore attractive for foreign investors. Since 1981 the SNTUC has pushed for skills upgrading and productivity improvements. It has even supported wage freezes and wage cuts during periods of recession.
In terms of Singapore's practices of industrial relations and collective bargaining, private sector collective bargaining usually takes place at the enterprise level. Generally speaking, collective bargaining covers two broad categories: matters related to salary and to conditions of employment.
Unions tend to conduct collective bargaining in a coordinated manner to achieve uniform results across different enterprises and categories of workers. Singapore's Industrial Relations Act forbids "rank and file" unions from engaging in collective bargaining on behalf of executive and managerial staff.