IV. Non-Oil Growth, Competitiveness, and the Labor Market41
69. The labor market in the United Arab Emirates is segmented in terms of sectors of employment, skills, and wages. It is estimated that Emirati nationals comprise about 10 percent of the overall labor force of 1.9 million, excluding military personnel. They are employed primarily in the public sector in mostly skilled jobs, and earn relatively high wages and social benefits. In contrast, expatriates make up most of the labor force, and are mainly employed in the private sector in mostly low- or semi-skilled jobs.
70. The abundant supply of expatriate workers at internationally competitive wages has been an important source of growth and competitiveness for the non-oil economy in the United Arab Emirates. 42, 43 In fact, real non-oil growth has averaged more than 7 percent per year from 1997-2002, and about 100,000 net new private sector jobs have been created on average each year over the same period. Most of these jobs have been low- and semi-skilled positions in sectors such as wholesale and retail trade, construction, manufacturing, and domestic household services. Only about 3,000 have been in skilled positions, in sectors such as banking and communications. At the same time, nearly 10,000 national workers have been entering the labor force each year in the last few years.
71. Given demographic dynamics, an increasing number of nationals are expected to enter the labor force over the medium term. Most of them are likely to seek employment in the private sector because the government is unlikely to remain the employer of first and last resort, except for professions such as teaching or in the health sector. In this context, the policy challenge facing the authorities is to expand private sector employment opportunities for nationals, while maintaining competitiveness and strong non-oil economic growth prospects.
72. This chapter, therefore, examines the relationship between the labor market and non-oil economic real growth in the United Arab Emirates, and studies policies to expand employment opportunities for nationals in the private sector. Section B analyzes the performance of non-hydrocarbon growth as well as the contribution of labor to non-oil growth over the last two decades. Section C focuses on the role of the labor market in maintaining competitiveness of non-oil exports despite an exogenous and rapid appreciation of the CPI-based real effective exchange rate in the second half of the 1990s. The last section develops an analytical model to propose policies for increasing private sector job opportunities for U.A.E. nationals.
B. Non-Oil Growth Performance
An overview of non-oil growth performance
73. The development of physical infrastructure in the 1970s and early 1980s set the base for rapid growth in services and manufacturing in the following years. Average annual growth of real non-oil GDP was about 7 percent in the 1990s through 2002, up from 3.6 percent in the previous decade. It was particularly strong in finance and insurance, wholesale and retail trade, and manufacturing, including petrochemicals. Tourism growth was also robust, especially in Dubai. As a result, the economy moved from being primarily oil-based to non-oil service based, with the non-oil share of GDP rising from close to 35 percent in 1980 to more than 70 percent in 2002 (Figure 6). 44
Figure 6.United Arab Emirates: Non-Oil Growth and the Changing Structure of GDP, 1980–2002
Sources: National authorities; and Fund staff estimates.
74. From the demand side, non-oil growth was mostly private consumption-led. This reflected the rise in income and population, particularly since the mid-1980s (Figure 7). In fact, per capita GDP in purchasing power parity terms increased by almost 66 percent, and the population grew by 150 percent from 1986 through 2002. However, non-oil investment also remained relatively high throughout this period, equivalent to more than 20 percent of GDP, reflecting in part the establishment and maturing of free trade zones.
Figure 7.United Arab Emirates: GDP Per Capita, Population, Consumption, and Investment, 1980–2002
Sources: National authorities; and IMF, World Economic Outlook.
Contribution of labor to non-oil growth
75. As a result of the open border foreign labor policy, employment growth contributed significantly to non-oil growth in the United Arab Emirates. This is confirmed by a growth accounting exercise, which decomposes non-oil output growth into the contributions of its constituent factor inputs, that is, labor, capital, and a residual, usually interpreted as total factor productivity (TFP). The exercise shows that, for plausible parameter assumptions, labor growth accounted for nearly one-third of non-oil growth in the 1980s, and more than one-half in the 1990s. 45
76. Access to expatriate workers at internationally competitive wages also contributed to avoiding the consequences of a “Dutch disease” usually observed in oil (or other natural resource) rich economies like the United Arab Emirates. 46 In most of these economies, the formal or organized labor market comprises mainly nationals. Thus, when the world price of oil rises, the wages of these nationals increase, making the non-oil export sector less competitive, and encouraging the adoption of import substitution policies. In contrast, the U.A.E. economy has been able to avoid “this disease” because it faces a highly elastic supply of foreign labor at competitive international wages and flexible contracts. In fact, the exogenous decline in real and relative wages of low-skilled expatriate workers over the 1990s contributed to boosting employment and output growth by lowering relative labor costs. As a result, employment of low-skilled workers was especially strong in that period, as was the growth of sectors such as trade and construction that use low-skilled workers more intensively (Figure 8). 47
Figure 8.United Arab Emirates: Employment, Wages, and Labor Productivity, 1981–2001
Sources: National authorities; and IMF staff estimates.
C. Non-Oil Exports and Competitiveness
77. Despite a sharp appreciation of the consumer price index (CPI)-based real effective exchange rate (REER) from the mid 1990s through 2002, non-oil exports and re-exports experienced rapid growth. The CPI-based REER appreciated by more than 30 percent owing to the nominal appreciation of the U.S. dollar in international markets, to which the U.A.E. dirham is pegged. 48 At the same time, non-oil export volume grew at over 16 percent per year, and re-exports at about 6 percent per year. Consequently, the United Arab Emirates’ share of the non-oil import market of its main trading partners doubled, and non-hydrocarbon and re-exports have become as large in value terms as hydrocarbon exports (Figure 9).
Figure 9.United Arab Emirates: CPI-based Effective Exchange Rates and Export Performance, 1980–2002
Sources: National authorities; and IMF staff estimates.
78. Competitiveness of non-oil exports derives partly from access to expatriate workers at international wages. Estimates of the unit labor cost-based REER, computed on the basis of trade with advanced country partners showed no appreciation or loss of competitiveness over the period (Figure 10). 49 Two other factors were important to maintain competitiveness. First, U.S. dollar appreciation did not hinder competitiveness vis-à-vis other partner countries in the region that also peg their currencies to the U.S. dollar. In fact, about 30 percent of non-oil and re-exports go to other GCC countries and Iran. In fact, using non-oil and re-export weights, the CPI-based REER has appreciated by less than 10 percent since the mid-1990s (Figure 10).
Figure 10.United Arab Emirates: Alternative Measures of Non-Oil Competitiveness, 1990–2002
(Index 1993 = 100)
Source: IMF staff estimates.
1/ Computed vis-à-vis advanced country partners.
79. The business-friendly environment and good infrastructure have also been key sources to maintain competitiveness and growth in the non-oil sectors. These factors have made the United Arab Emirates an attractive place to invest in, particularly in the free zones, for companies seeking to serve the growing markets of the Middle East, North Africa, India, and Pakistan.
D. Policies for the Employment of Nationals
80. With nearly 50 percent of the national Emirati population currently below the age of 15, a growing number of nationals will certainly enter the labor force in the period ahead. They will most probably seek employment in the private sector because the government is unlikely to remain the employer of first and last resort. In this regard, increasing private sector employment opportunities for nationals may come from two sources. 50 A larger number of skilled jobs should be created in the coming years as the United Arab Emirates seeks to become a technology and service hub within the region, and as investment increases in knowledge-intensive sectors, such as in the news media, electronics, and internet free trade zones in Dubai. Nevertheless, most job opportunities are likely to come from replacing expatriates in existing skilled positions. So, what policies can the authorities implement to increase employment opportunities for nationals while maintaining competitiveness and strong non-oil growth prospects? To answer this question, consider the following model.
A model of employment and labor substitution
81. A composite non-oil output good is produced in the private sector using three types of labor—nationals, skilled expatriates, and low-skilled expatriates. Nationals and skilled expatriates are assumed to be substitutes in production, which is given by:
where A is the level of technology or total factor productivity (TFP), K is the capital stock, L is the number of low-skilled expatriate workers, H is the number of skilled expatriate workers, N is the number of nationals, h is the human capital or efficiency of national workers, σ and is the elasticity of substitution between skilled expatriate workers and nationals. A high value of σ indicates a high degree of substitutability.
82. Private sector demand for national labor, N d, depends on the costs of employing the different types of workers, the human capital or efficiency of national workers, the level of technology, and the stock of capital, in the following manner: 51
where w and β are, respectively, the private sector wages and benefits paid to nationals, c is the relative cost of firing them, wL is the compensation of low-skilled expatriates, wH is the compensation of skilled expatriates, and µ is a constant. As the different factors change, demand changes:
(i) As expected, demand is inversely related to the cost of hiring and firing national workers.
(ii) It is also inversely related to the cost of employing low-skilled expatriate workers. Since a reduction in the number of low-skilled employees reduces the marginal productivity of national workers, an increase in the cost of employing low-skilled workers reduces demand not only for these workers but also for national workers.
(iii) If skilled expatriate workers are sufficiently substitutable with nationals, then demand is positively related to the cost of employing skilled expatriates because increases in these costs lead firms to reduce their demand for skilled expatriate workers and to substitute them with national workers. 52
(iv) It is also positively related to the level of human capital or efficiency of national workers, the stock of capital, and technology, because increases in these factors raises the return to employing an additional national worker.
83. The supply of national workers to the private sector depends on the compensation relative to their reservation wages. Reservation wages in turn are determined by the wage, social benefit, and employment policies of the government. If the public sector provides relatively high wages, social benefits, and job security to nationals, then the reservation wage and benefits, w*+β*, will be relatively high. For simplicity, assume an elastic supply of nationals at the reservation wage:
84. Based on equation (4), several policies can be adopted to increase job opportunities for nationals in the private sector. These may be grouped into three broad categories depending on whether they directly affect wages and employment costs, acquisition of human capital, and investment in capital and technology. These policies are structural and market-based, in contrast to quantitative measures, such as job quotas for nationals in the private sector, which may adversely affect competitiveness and non-oil growth by raising costs and limiting employment flexibility. 53
85. Reducing the relatively high wages in the public sector is likely to lower the reservation wage and increase the willingness of nationals to acquire skills or human capital valuable to private sector employers. Announcing and enforcing strict limits on public sector hiring is likely to further lower the reservation wage by decreasing the likelihood that the public sector will act as the employer of first and last resort for nationals.
86. Separating wages and social benefits in the public sector and providing benefits to all working nationals—not just to those employed in the public sector—is likely to reduce the incentive for nationals to seek public sector employment and lower the reservation wage. Moreover, giving time-specific subsidies for the employment of nationals will likely increase private sector demand for them by lowering their employment costs. However, the fiscal cost will need to be less than the employment benefit for the policy to be welfare improving. 54
87. Relatively higher firing costs for nationals, including lengthy appeal and investigation of dismissals, raises their relative cost of employment, thus reducing demand. Therefore, establishing a clear set of rules for the appeal of dismissals, including fines or penalties associated with wrongful dismissals, and a mechanism for the rapid resolution of appeals is likely to lower the relative cost of employing nationals, and reduce the disparity in labor mobility between nationals and expatriate workers across sectors.
88. If skilled expatriate workers are sufficiently substitutable with national workers, then increasing the relative cost of hiring skilled expatriate workers, such as by imposing income taxes or fees on the employment of skilled expatriates, is likely to result in substitution away from skilled expatriates and towards nationals. 55 Such taxes or fees should only be applied to skilled expatriates and not to low-skilled expatriates because the latter complement the productivity of nationals. If taxes or fees are applied to low-skilled expatriate workers, then the cost of hiring them will increase, and the demand for national workers is likely to decrease.
89. Enhancing the human capital of nationals and the acquisition of skills that are valuable to the private sector is likely to increase demand and employment. To this end, providing education, including vocational training, reforming school curricula, encouraging firms to establish internships, awarding scholarships as well as targeted training vouchers, and fostering self employment will likely build necessary skills and expertise among prospective national workers.
Investment in capital and knowledge
90. Facilitating the adoption of new technologies and the accumulation of capital is also likely to increase private sector demand and employment by raising the productivity of nationals. In this regard, the establishment of knowledge- and technology-intensive free trade zones in Dubai is likely to help. Moreover, continued out-sourcing of government services and extending 100 percent foreign ownership into non-free trade zone areas will likely promote competition and improve resource allocation, leading to higher productivity and investment.
E. Summary and Conclusions
91. The non-oil output and employment performance of the United Arab Emirates have been very strong, particularly in the last decade, with the abundant supply of expatriate workers at internationally competitive wages being a key source of non-oil growth and competitiveness. In fact, the number of net new jobs created by the private sector has been about 10 times the number of nationals entering the work force each year. However, most of these jobs have been in low- and semi-skilled positions, which have not attracted skilled nationals. As a result, the majority of nationals have continued to find employment in the government.
92. With nearly 50 percent of the national population below the age of 15, a rapidly growing number of nationals are expected to enter the labor force in the period ahead. Since the public sector is not likely to remain the employer of first and last resort, nationals will increasingly need to seek employment in the private sector. The policy challenge therefore is to expand job opportunities for nationals while maintaining non-oil growth and competitiveness.
An effective employment strategy for nationals may include:
- Increasing the relative attractiveness of working in the private sector by lowering the wage differential between the public and private sectors; making social benefits available to all working nationals, irrespective of sector of employment; and announcing strict limits on public sector employment.
- Reducing disparities in labor mobility by creating a level playing field between hiring and firing national workers vis-à-vis expatriate workers.
- Encouraging skill acquisition among nationals by strengthening educational and vocational training, providing time-specific incentives, such as subsidies and scholarships, and promoting self-employment.
- Facilitating improvements in productivity and investment in capital by out-sourcing government services, and extending majority foreign ownership in non-free trade zone areas.
- Using price-based rather than quantity-based market interventions by charging, for instance, fees for employing or income taxes on skilled expatriate workers.
93. In sum, adopting structural and market-based policies to increase private sector employability and job opportunities for nationals will also maintain the labor market flexibility that is key for sustaining strong non-oil growth and competitiveness.
94. A growth accounting exercise decomposes non-oil output growth into the contributions of its constituent factor inputs, that is, labor, capital, and a residual factor, usually interpreted as total factor productivity (TFP). For the United Arab Emirates, growth in labor employment and capital account for nearly all non-oil output growth.
95. Consider the standard Cobb-Douglas production function for non-oil output, Y = K α (AL) l-α, where Y is output, K is the capital stock of the non-oil sector, L is labor input into the non-oil sector, A is the level of TFF or technology, and α is the capital share of income. Taking logarithms and differentiating with respect to time yields the decomposition of output growth: gY = αgK + (1-α)gL + (l-α)gA, where g x is the growth rate of variable x, α gK is the contribution of capital to non-oil output growth, (l-α)gL is labor’s contribution, (l-α)gA is the residual or the contribution of TFP growth, and (1-α) is labor’s share of income.
96. Data available suggest that worker compensation is about 40 percent of income. The number is low by international standards, but given that the vast majority of workers are expatriates and that employers or owners earn large markups, it is reasonable to assume a value of 0.6 for capital’s share, α.
97. Labor is proxied for by the number of employees since the number of hours worked is not available. The stock of capital is constructed using the perpetual inventory method. In other words, capital accumulation is given by Kt+1 = (1-δt) Kt + It where Kt is capital at time t, It is investment, and δt is the depreciation rate. Given the high replacement rate of structures in the United Arab Emirates and the increasing share of equipment in the stock of capital, δt is assumed to be 10 percent for the 1980s and 12 percent for the 1990s. 56 Finally, an initial capital-output ratio of 2.1 is assumed for 1980. 57 These assumptions yield the following results:
(Average In the period, in percent)
98. Growth in labor and capital account for almost all non-oil output growth. Labor’s contribution, given by(l-α)gL, was about 30 percent for the 1980s and over 50 percent for the 1990s, confirming that the abundant supply of expatriate labor has been a major engine of output growth. Capital accumulation accounted for most of the rest. TFP’s contribution was about 3 percent, which is similar to numbers in other developing countries. 58
99. Let private sector non-oil output, Y, be given by:
where A is the level of technology or TFP, K is the capital stock, L is the number of low-skilled expatriate workers, H is the number of skilled expatriate workers, N is the number of nationals, h is the quality or efficiency of national workers, and σ is the elasticity of substitution between skilled expatriate workers and nationals. A high value of σ indicates a high degree of substitutability; σ=∞ corresponds to perfect substitutability, while σ=0 corresponds to zero or lack of substitutability between skilled expatriate workers and nationals. Since nationals are better substitutes for skilled expatriates than for low-skilled expatriates, the elasticity of substitution between nationals and skilled expatriates is higher than between nationals and low-skilled expatriates. From the Cobb-Douglas production function, the elasticity of between nationals (or skilled expatriates) and low-skilled expatriates is 1. Therefore, σ is greater than 1.
Each period, a firm chooses different types of labor to maximize profits, given wages, technology, and capital. Workers are hired on a period-by-period basis.
where w and β are respectively the real wage and benefits paid to a national worker, wL is the real wage of a low-skilled expatriate worker, wH is the real wage of a skilled expatriate worker, and c is the cost associated with firing or separating from a national worker. Note that wL and wH are the international wages for low-skilled and skilled workers respectively, and hence are exogenously given.
Private sector demand for national labor is obtained by solving the first order conditions and re-writing in terms of the exogenous variables:
This chapter was written by Rishi Goyal (MED).
Oil or hydrocarbons refer to crude oil and associated natural gas. Refined products are included in the non-oil or non-hydrocarbon sector.
Bangladesh, India, and Pakistan supply the majority of expatriate workers, though large numbers also come from Indonesia and the Philippines.
Oil output growth has varied significantly over the years in line with OPEC-mandated changes in quota.
Capital accumulation accounted for the remainder. The contribution of TFP growth was negligible, similar to the experience of other developing countries. Details are provided in Appendix I.
The Dutch disease refers to the negative output and employment effects of an oil (or natural resource) boom on non-oil exports; thus, leading to a contraction in the country’s tradable sector.
Evidence for declining real and relative wages for low-skilled workers comes from labor productivity data, since wage data by skill type are not available for the United Arab Emirates. Labor productivity in the non-oil sector is defined as the ratio of real value-added in the non-oil sector to the number of workers employed. Labor productivity declined in sectors such as construction and trade that employ low-skilled workers more intensively, and rose in sectors such as finance, insurance, and communications that employ skilled workers more intensively. This mirrors a similar phenomenon in several other countries over the last two decades.
Computed by the Information Notice System (INS), the weights used in the standard CPI-based REER calculations are based on trade in oil, manufactured goods, and others vis-à-vis developed and developing countries.
Unit labor cost (ULC), or the labor cost to produce one unit of output, is obtained by dividing nominal compensation by average real labor productivity. The ULC-based REER is estimated for the business sector, defined as the non-oil sector, excluding government services, and is computed vis-à-vis advanced country partners for which ULC data is available. These countries account for over 70 percent of total exports.
The unemployment rate among nationals was officially estimated at less than 3 percent in 1995, the year of the last population census.
That is, if σ > 1+(1-ɑ)/γ> 1, then demand for nationals rises with increases in wH. For the United Arab Emirates, the upper bound for 1+(1-ɑ)/γ estimated to be 1.5.
The U.A.E. government has thus far adopted quotas only in the banking sector, albeit applied flexibly.
These subsidies could be financed, for instance, by fees or income taxes on skilled expatriate workers.
Assuming that the employer in the United Arab Emirates will cover the cost of paying the income tax of his expatriate workers.
These depreciation rates imply replacement within two decades, which correspond to the rapid replacement rate of structures and equipment in the United Arab Emirates.
The steady state capital output ratio is usually 3. Basic infrastructure investments were ongoing in 1980 and, given that infrastructure investment precedes investment in other sectors, an initial value of 2.1 was assumed.
The qualitative result of labor and capital accounting for almost all non-oil output growth is robust to the initial capital stock. The quantitative results for the 1980s are sensitive to the initial value, given the short available time series of investment data. The quantitative results for the 1990s are more robust, especially given the assumption of high depreciation rates.