Abstract
This chapter reviews developments in GDP over the past several decades. The analysis shows that accumulation of labor and capital explains the bulk of overall output growth since 1990, with changes in total factor productivity playing only a minor role. Moreover, while increases in total factor productivity (TFP) during 1990-2009 have been close to the worldwide average, the pace of TFP growth fell during the 2000s. This suggests scope for increasing the efficiency of factor markets and highlights the importance of recent reforms to promote knowledge-based activity.
I. Realizing Growth Objectives: Transitioning from Factor Accumulation to Productivity Improvement 1
This chapter reviews developments in GDP over the past several decades. The analysis shows that accumulation of labor and capital explains the bulk of overall output growth since 1990, with changes in total factor productivity playing only a minor role. Moreover, while increases in total factor productivity (TFP) during 1990-2009 have been close to the worldwide average, the pace of TFP growth fell during the 2000s. This suggests scope for increasing the efficiency of factor markets and highlights the importance of recent reforms to promote knowledge-based activity.
A. Introduction
1. The oil-driven booms and busts of the 1970s and 1980s have, over the past two decades, given way to comparatively steady growth in the Saudi economy. Activity in the non-oil sector has been particularly robust and has in recent years been accounting for a growing share of real GDP. At the same time, the economy has absorbed large numbers of workers and investment has surged. This chapter reviews Saudi Arabia’s growth record, contrasts developments in the non-oil sector to the economy as a whole, and evaluates the extent to which increases in output can be attributed to different factors of production.
2. Using a growth accounting framework, changes in output are decomposed into contributions from physical and human capital as well as from the productivity with which those inputs are used. The findings suggest that output growth over 1990–2010 can largely be attributed to factor accumulation, with little contribution from total factor productivity (TFP). While the applied framework does not explain why productivity has evolved as it has, the analysis highlights areas with likely scope for increasing efficiency and, hence, raising the amount of output for a given level of input. It is found that output has been particularly intensive in labor, which suggests that steps to improve the efficiency of labor markets have substantial potential to improve economic outcomes.
3. While growth accounting has generated a vast literature, only a small number of previous studies have covered Saudi Arabia. Among these are Baier et al. (2006) who in a large cross-country investigation find that growth in TFP per worker over the past several decades has been negative in Saudi Arabia, something they also find for almost all other countries in the Middle East. Focusing on the Middle East and North Africa (MENA) region, Bisat et al. (1997), IMF (2011), Keller and Nabli (2002), and Sala-i-Martin and Artadi (2002), get similar results.
4. Investigations into TFP growth in Saudi Arabia’s non-oil economy are even more scarce. Difficulties in obtaining separate data for the non-oil sector in sufficiently long time series have been a key obstacle. Among the few studies we are aware of that apply growth accounting to the non-oil sector, Bisat et al. (1997) find that TFP growth was strongly positive during 1975–84 but then turned negative in 1986–95. This is consistent with Al-Khatib (2011) who finds that average TFP growth in the non-oil economy was positive for most of the 1970s, mainly negative in the 1980s and 1990s, and then positive again during 2000–07, with a near-zero average annual growth rate over the whole period. The present analysis complements these two studies by extending the investigation period to 2010 and contrasting the results for Saudi Arabia with those for other countries, using a consistent international framework.
5. The rest of this chapter proceeds with a review of trends in GDP and its underlying inputs over the past several decades. The following section then applies growth accounting to decompose changes in both total and non-oil GDP during 1990–2009 into its sources. This is done using the longest time series available from official sources, and results are compared to an extensive set of international comparators. The chapter concludes with a review of growth in new industries and a discussion of the importance of economic diversification for continued increases in income.
B. Trends in GDP and its Inputs
6. Growth in Saudi Arabia’s economy has to a large extent been driven by developments in the oil sector. As shown in Figure I.1, the spike in oil prices in the 1970s was accompanied by a surge in oil production as well as by high real GDP growth rates averaging over 10 percent a year. This process was reversed in the early 1980s when falling oil prices led to a sharp decline in oil output and negative GDP growth. After abandoning its role as the swing producer, Saudi Arabia gradually increased oil output again in the second half of the 1980s. Since then, oil production has been more stable, but with some increase over the past decade bringing production back close to the peak level of 1980. The result has been greater economic stability, with decade-average real GDP growth holding broadly steady at about 3⅓ percent a year during the 1990s and 2000s although with still considerable volatility from year to year.
Saudi Arabia: Trends in GDP and Factor Inputs, 1970–2011
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: World Economic Outlook Database; BP Statistical Review; CDSI; Barro-Lee dataset; and IMF staff calculations.Saudi Arabia: Trends in GDP and Factor Inputs, 1970–2011
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: World Economic Outlook Database; BP Statistical Review; CDSI; Barro-Lee dataset; and IMF staff calculations.Saudi Arabia: Trends in GDP and Factor Inputs, 1970–2011
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: World Economic Outlook Database; BP Statistical Review; CDSI; Barro-Lee dataset; and IMF staff calculations.7. Non-oil activity has increasingly been contributing to output, although the economy remains highly oil-dependent. Steadier than overall GDP, non-oil GDP growth has been gradually rising over the past two decades, going from an average rate of 2.7 percent a year in the 1990s to 4.3 percent during 2000–11. As a result, the oil sector’s share in total real GDP has declined from almost 40 percent in 1991 to less than 30 percent in 2011. This does not mean, however, that the economy has become less dependent on oil, as higher oil prices caused steep increases in oil revenue over the past decade. Indeed, in nominal terms, the oil sector’s contribution to GDP at current prices surged from about 30 percent in the late 1990s to currently close to 60 percent.
8. From an international perspective, overall economic growth in Saudi Arabia after the 1970s boom has been middling, while non-oil growth has in recent years been comparatively strong. Total real GDP growth since the 1990s has been close to the worldwide median—lower than that of the fastest-growing emerging and developing economies, broadly on par with Latin America and the Caribbean, and higher than in advanced economies (Figure I.2). Non-oil growth during the past decade has fared better, outperforming the worldwide median by, on average, about a third of a percentage point a year. Taking into account the rapid increase in population, however, growth has been disappointing, with total real GDP per capita currently about 40 percent lower than in 1980 and non-oil GDP per capita about 20 percent lower.
Real GDP Growth, 1970–2009
(Percent annual change, period average)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: IMF World Economic Outlook database; World Bank, World Development Indicators; and IMF staff calculations.Real GDP Growth, 1970–2009
(Percent annual change, period average)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: IMF World Economic Outlook database; World Bank, World Development Indicators; and IMF staff calculations.Real GDP Growth, 1970–2009
(Percent annual change, period average)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: IMF World Economic Outlook database; World Bank, World Development Indicators; and IMF staff calculations.9. Looking at factor inputs, investment levels surged in the 1970s but then fell back in the early 1980s before increasing again in the past decade. From 1999 to 2009 the ratios of total investment to GDP and of non-oil investment to non-oil GDP both increased by almost 15 percentage points to, respectively, 35 and 40 percent. Most of this increase can be attributed to government investments in the non-oil sector, which increased by 14 percentage points of non-oil GDP in the 10 years to 2009 (Figure I.3). These investments were made possible by higher oil revenue and have been channeled into infrastructure as well as social and education spending, following priorities set out in successive national development plans.
Non-oil Investment, 1985–2011
(Percent of non-oil GDP, at constant prices)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: Central Department of Statistics and Information (CDSI); and IMF staff estimates.Non-oil Investment, 1985–2011
(Percent of non-oil GDP, at constant prices)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: Central Department of Statistics and Information (CDSI); and IMF staff estimates.Non-oil Investment, 1985–2011
(Percent of non-oil GDP, at constant prices)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: Central Department of Statistics and Information (CDSI); and IMF staff estimates.10. Increases in labor inputs have been even more substantial. Tracking developments in GDP, overall employment growth averaged about 8 percent a year during the 1970s before declining to about 2 percent in the 1990s and then picking up to almost 3 percent in the 2000s. These figures are broadly mirrored in the working-age population, which is estimated to have risen much more rapidly than in other regions (Figure I.4). In addition, there has been major change in the composition of labor inputs. During the 1970s and 1980s employment growth was largely driven by non-Saudi workers, who by the end of that period accounted for about two thirds of total employment. In the last two decades, however, the growth rate of Saudi employment has been higher than that of non-Saudis, although the still larger base of non-Saudi workers has meant that the absolute increase in Saudi and non-Saudi employment has recently been roughly the same. In addition to the higher numbers, the workforce has also become much more educated: more than 80 percent of the working-age population is estimated to have had at most a primary school education in 1970, but by 2010 this proportion had dropped below 40 percent.
Working-Age Population Growth, 1970–2009
(Percent annual change, ages 15-64)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: World Bank, World Development Indicators.Working-Age Population Growth, 1970–2009
(Percent annual change, ages 15-64)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: World Bank, World Development Indicators.Working-Age Population Growth, 1970–2009
(Percent annual change, ages 15-64)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: World Bank, World Development Indicators.C. Growth Accounting
11. The contribution of the different factors of production to GDP can be derived in a standard growth accounting framework. 2 This assumes the following general production function:
where output (Y) is a function of two production factors, human capital (H) and physical capital (K), and that function is augmented by total factor productivity (A). Here, human capital, H = LQ, is the product of raw labor (L) and labor quality (Q). Assuming constant returns to scale in production and perfectly competitive factor markets, the production function can be transformed into:
where vL is the labor cost share in production. This means that the percent change in total factor productivity can be calculated as the percent change in output less the cost-weighted average of the percent changes in the factors of production.
12. The main challenge in implementing the growth accounting framework is to obtain accurate data. The Total Economy Database (TED) made available by the Conference Board represents perhaps the most comprehensive effort to date towards integrating international data into one internally consistent database.3 It includes series decomposing GDP growth into its sources, and covers over 120 countries going back to 1990.4 We follow the same methodology as used in TED, but recalculate the results for Saudi Arabia using series for GDP, employment, and investment from the Central Department of Statistics and Information (CDSI). Moreover, for Saudi Arabia we extend the calculation to also cover the non-oil economy, using CDSI’s time series for non-oil investment and employment.
13. Data limitations remain a key constraint. One important element of uncertainty pertains to the labor cost share, νL. This figure may be derived from national accounts and can vary over time, but those data are often not available. Indeed, for Saudi Arabia as well as for many other non-OECD economies, TED just assumes a constant labor cost share of 0.5.5 Another important limitation relates to the measurement of human capital, H. TED’s series for raw labor, L, refers to hours worked; but if those data are not available it is measured as employment or, in some cases, the size of the working-age population. Changes in labor quality, Q, are particularly hard to measure, with estimates generally based on educational attainment and thus not accounting for changes in the value of that education. To reflect data uncertainty, we conduct sensitivity analysis on key parameters.
14. Following TED, changes in labor quality are calculated using a Törnquist index based on the shares of labor in low, medium, and high skill groupings and with relative wages as the respective weights.6
where
Here, vi,t is the share in labor compensation of labor type i at time t, qi is the share of type i labor, ωi is the wage of type i labor, and
15. Finally, the physical capital stock, K, is constructed from data on investment, I, using the perpetual inventory method:
where the initial capital stock, K0, is calculated as K0 = I0/(δ+g) based on a constant depreciation rate, δ, and an initial output growth rate, g.7
Results
16. The growth accounting results indicate that overall GDP for Saudi Arabia has in the past two decades been mainly driven by factor accumulation. As shown in Figure I.5, of the 3.2 percent average annual rate of real GDP growth during 1990–2009, the decomposition reveals that 1.5 percentage points can be attributed to accumulation of physical capital. Another 1.5 percentage points can be attributed to increasing human capital, most of it from growing employment and a smaller part from the change in composition toward higher education levels. As a result, overall TFP growth is found to have been just marginally positive with a period average rate of 0.2 percent a year.8 On an annual basis, however, the results show considerable variation, with large positive TFP growth in some years. This was notably the case in the early 1990s, reflecting that higher oil production in those years was achieved with relatively small increases in labor and capital. Nevertheless, for most years since 1990, TFP growth is found to have been negative.
Saudi Arabia: Growth Decomposition, 1970–2010
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: The Conference Board Total Economy Database; CDSI; and IMF staff estimates.Saudi Arabia: Growth Decomposition, 1970–2010
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: The Conference Board Total Economy Database; CDSI; and IMF staff estimates.Saudi Arabia: Growth Decomposition, 1970–2010
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: The Conference Board Total Economy Database; CDSI; and IMF staff estimates.17. The decomposition of non-oil GDP indicates that TFP growth in that sector has on average been somewhat higher than for the economy as a whole. For 1990–2009, the average growth rate of non-oil GDP at 3.4 percent was 0.2 percentage points higher than for total GDP. Nevertheless, the results show that that factor inputs had a slightly lower contribution to non-oil growth than they did for total GDP. As a result, average TFP growth for the non-oil economy is found to have been higher at 0.5 percent a year.
18. In both the economy as a whole and the non-oil sector, over the past two decades the pace of factor accumulation has generally been increasing and TFP growth falling. Reflecting the surge in investment, physical capital’s contribution to both overall and non-oil GDP growth more than doubled from the 1990s to the 2000s. The contribution from human capital also increased, mainly as a result of raw employment numbers. For the economy as a whole, the increase in factor accumulation coincided with almost flat GDP growth, leading to a considerable reduction in TFP. For the non-oil economy, however, GDP growth also accelerated, leaving only a slight decline in TFP growth.
19. Although modest, the estimates of Saudi Arabia’s TFP growth are not far from the international average. Indeed, as shown in Figure I.6, low TFP growth is not unusual, with about one third of the countries in TED displaying negative TFP growth over 1990–2009, although in most cases by less than half a percentage point per year.9 Among G-20 economies (Figure I.7), South Korea and China top the chart with average TFP growth of about 2½ percent a year, while eight countries have had negative TFP growth. For Saudi Arabia, the relative performance of TFP follows from well-above-average growth in labor input—for both raw labor and labor quality—coupled with somewhat below-average growth in physical capital and near-average growth in GDP.
Cross-Country Growth Patterns, 1970–2009
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: The Conference Board Total Economy Database; CDSI; and IMF staff estimates.Cross-Country Growth Patterns, 1970–2009
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: The Conference Board Total Economy Database; CDSI; and IMF staff estimates.Cross-Country Growth Patterns, 1970–2009
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: The Conference Board Total Economy Database; CDSI; and IMF staff estimates.
TFP Growth, 1990–2009
(Annual percent change, period average)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: The Conference Board Total Economy Database; CDSI; and IMF staff estimates.TFP Growth, 1990–2009
(Annual percent change, period average)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: The Conference Board Total Economy Database; CDSI; and IMF staff estimates.TFP Growth, 1990–2009
(Annual percent change, period average)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Sources: The Conference Board Total Economy Database; CDSI; and IMF staff estimates.20. Low TFP growth suggests that the factors of production are not being used as efficiently as possible. Continued technological advancements should tend to raise productivity over time. For oil producers, however, part of the story behind low TFP growth may be that natural resource extraction naturally involves increasingly more input for each unit of output since the resources that are easiest to access will tend to be extracted first. Low TFP growth in the oil sector could also be a consequence of a greater share of the output being sold domestically at low prices, which would reduce measured value added in the sector and offset positive effects from technological improvements. Elsewhere, low or negative TFP growth may be the result of conflict or of countries shifting resources into relatively low-productivity areas. Shortcomings in the data may also be a factor, with particular uncertainty surrounding the hard-to-measure concept of human capital—an especially important issue for Saudi Arabia given the substantial changes in labor force composition that have occurred over recent decades in terms of nationality and average education levels. In some countries, not least Saudi Arabia where increased activity in the non-oil private sector may not be fully captured, low TFP could also be an indication of underestimated GDP growth.
21. Uncertainties about the exact level of TFP growth notwithstanding, the growth accounting results indicate scope for raising economic growth in Saudi Arabia. There is clear evidence that GDP growth has been driven mainly by factor accumulation and this implies room for deriving more output from the same level of inputs. In particular, Saudi Arabia’s comparatively large input of labor could have been expected to generate more GDP. That this has not been the case suggests that raising the efficiency with which the labor market operates has substantial potential for improving economic outcomes, as discussed in the next chapter. At the same time, for labor inputs to play a greater role and to absorb the growing population, there must also be demand for labor, which requires continued progress on economic diversification.
D. Sectoral Reallocation, Economic Diversification, and the Demands Ahead
22. Sustaining growth in income and employment hinges on expansion of the non-oil economy. Endeavors in this direction have been a central part of the country’s development plans, which have increasingly been emphasizing economic diversification and transition toward knowledge-based forms of production as central objectives. Education is high on the agenda, with recent steps including the opening of several major universities and a sharp increase in the number of scholarships provided for study abroad. There has also been a strong focus on stimulating private enterprise. Indeed, efforts to improve the business environment have led to marked increases in the country’s ranking in the World Bank’s Doing Business Report (12th out of 183 in the 2012 report) and the World Economic Forum’s World Competitiveness Report (17th out of 142 in the 2011–12 report).
23. The area where the transition toward new types of production has been most pronounced has been in the downstream processing of oil and natural gas feedstock into petrochemicals and plastics. Real GDP from petrochemical and other non-oil manufacturing has grown at an average of almost 7 percent a year since 1980—faster than any other sector of the economy, even as services and construction have picked up (Figure I.8). According to the Saudi Industrial Development Fund, between 1974 and 2010 the number of producing units in the industrial sector increased from less than 200 to more than 4,500, while the stock of capital invested in the sector grew to over 50 percent of non-oil GDP.
Growth in Non-oil GDP, 1980–2009
(Percent annual change of GDP at constant prices)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: CDSI.Growth in Non-oil GDP, 1980–2009
(Percent annual change of GDP at constant prices)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: CDSI.Growth in Non-oil GDP, 1980–2009
(Percent annual change of GDP at constant prices)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: CDSI.24. Although crude oil remains by far the largest source of export receipts, the expansion of non-oil manufacturing has helped diversify revenue streams. Apart from food processing, which is primarily for local consumption, non-oil manufacturing is dominated by production of intermediate inputs used in factories abroad. Reflecting the increase in production, exports of chemicals, plastics, and metals have accordingly surged from under 2 percent of nominal non-oil GDP in 1985 to over 10 percent in 2010 (Figure I.9). The share of these non-oil exports in total external receipts has not increased as much, but that largely reflects the impact of higher oil prices on total export values.
Non-oil Exports, 1985–2010
(Percent of non-oil GDP at current prices)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Non-oil Exports, 1985–2010
(Percent of non-oil GDP at current prices)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Non-oil Exports, 1985–2010
(Percent of non-oil GDP at current prices)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Output and Employment by Sector, 1989–2009
Includes Agriculture, forestry & fishing; Non-oil mining; Electricity, gas & water; and Government services.
Output and Employment by Sector, 1989–2009
| Oil sector | Non-oil manufacturing | Construction | Non-gov. services | Other non-oil1 | Total non-oil economy | Total economy | ||
|---|---|---|---|---|---|---|---|---|
| GDP (in billions of SAR at constant 1999 prices) | ||||||||
| 1989 | 131 | 24 | 32 | 129 | 124 | 308 | 439 | |
| 1999 | 191 | 45 | 39 | 157 | 162 | 403 | 594 | |
| 2009 | 219 | 83 | 58 | 257 | 210 | 608 | 827 | |
| Employment (in 1000s) | ||||||||
| 1989 | 54 | 415 | 800 | 2,714 | 1,071 | 5,000 | 5,054 | |
| 1999 | 97 | 484 | 653 | 3,503 | 1,433 | 6,073 | 6,170 | |
| 2009 | 91 | 505 | 965 | 4,937 | 1,650 | 8,057 | 8,148 | |
| Share in total employment (in percent) | ||||||||
| 1989 | 1.1 | 8.2 | 15.8 | 53.7 | 21.2 | 98.9 | 100.0 | |
| 1999 | 1.6 | 7.8 | 10.6 | 56.8 | 23.2 | 98.4 | 100.0 | |
| 2009 | 1.1 | 6.2 | 11.8 | 60.6 | 20.3 | 98.9 | 100.0 | |
| GDP per employee (in 1000s of SAR at constant 1999 prices) | ||||||||
| 1989 | 2,417 | 57 | 40 | 48 | 115 | 62 | 87 | |
| 1999 | 1,962 | 92 | 60 | 45 | 113 | 66 | 96 | |
| 2009 | 2,403 | 165 | 60 | 52 | 127 | 76 | 102 | |
| Average annual increase in labor productivity (in percent) | ||||||||
| 1990s | −1.6 | 5.1 | 4.6 | −0.5 | −0.2 | 0.8 | 1.2 | |
| 2000s | 2.6 | 6.3 | 0.3 | 1.6 | 1.3 | 1.3 | 0.6 | |
Includes Agriculture, forestry & fishing; Non-oil mining; Electricity, gas & water; and Government services.
Output and Employment by Sector, 1989–2009
| Oil sector | Non-oil manufacturing | Construction | Non-gov. services | Other non-oil1 | Total non-oil economy | Total economy | ||
|---|---|---|---|---|---|---|---|---|
| GDP (in billions of SAR at constant 1999 prices) | ||||||||
| 1989 | 131 | 24 | 32 | 129 | 124 | 308 | 439 | |
| 1999 | 191 | 45 | 39 | 157 | 162 | 403 | 594 | |
| 2009 | 219 | 83 | 58 | 257 | 210 | 608 | 827 | |
| Employment (in 1000s) | ||||||||
| 1989 | 54 | 415 | 800 | 2,714 | 1,071 | 5,000 | 5,054 | |
| 1999 | 97 | 484 | 653 | 3,503 | 1,433 | 6,073 | 6,170 | |
| 2009 | 91 | 505 | 965 | 4,937 | 1,650 | 8,057 | 8,148 | |
| Share in total employment (in percent) | ||||||||
| 1989 | 1.1 | 8.2 | 15.8 | 53.7 | 21.2 | 98.9 | 100.0 | |
| 1999 | 1.6 | 7.8 | 10.6 | 56.8 | 23.2 | 98.4 | 100.0 | |
| 2009 | 1.1 | 6.2 | 11.8 | 60.6 | 20.3 | 98.9 | 100.0 | |
| GDP per employee (in 1000s of SAR at constant 1999 prices) | ||||||||
| 1989 | 2,417 | 57 | 40 | 48 | 115 | 62 | 87 | |
| 1999 | 1,962 | 92 | 60 | 45 | 113 | 66 | 96 | |
| 2009 | 2,403 | 165 | 60 | 52 | 127 | 76 | 102 | |
| Average annual increase in labor productivity (in percent) | ||||||||
| 1990s | −1.6 | 5.1 | 4.6 | −0.5 | −0.2 | 0.8 | 1.2 | |
| 2000s | 2.6 | 6.3 | 0.3 | 1.6 | 1.3 | 1.3 | 0.6 | |
Includes Agriculture, forestry & fishing; Non-oil mining; Electricity, gas & water; and Government services.
25. The expansion of non-oil manufacturing also helps to shed light on the observed productivity developments. While lack of comprehensive data on investment by sector preclude detailed growth accounting, employment data make it possible to calculate labor productivity at the sectoral level. As shown in Table I.1, the rapid pace of output growth in non-oil manufacturing has been achieved with relatively little additional employment: the sector’s share in total employment fell steadily from 8½ percent in the mid-1980s to 6.2 percent in 2009. This led to faster increases in labor productivty than in any other sector and helped support overall productivity growth. At the same time, because the share of total employment in non-oil manufacturing has fallen, labor has become more concentrated in areas of lower productivity, which has subtracted from overall productivity gains.
26. Effects of shifting employment patterns on productivity can be directly measured. Overall growth in labor productivity can be decomposed as follows:
where yi,t = Yi,t/Li,t is labor productivity in sector i at time t, and si,t = Li,t/Lt is the sector’s share in total employment. Here, the first term on the right is the direct effect due to sectoral productivity growth. The second term is the shift effect due to changes in employment shares. The final term is the cross effect that captures interaction among the different variables.
27. The decomposition shows that the greater concentration of labor in sectors with relatively low productivity has been a substantial drag on overall productivity growth during the past decade. As shown in Figure I.10, the shift effects have been considerable. With productivity in the oil sector 30–40 times higher than in the rest of the economy, the small increase in the share of employment in the oil sector during the 1990s had a major positive impact on labor productivity growth in that decade. This effect reversed itself in the 2000s when the share of employment in the oil sector fell back, leading to a reduction in overall labor productivity, despite most individual sectors recording advances in productivity growth. Within the non-oil economy, shift effects stemming from the declining share of employment in manufacturing, and to a lesser extent the increasing share of employment in construction, were also a drag on productivity growth in the 2000s.
Decomposition of Labor Productivity, 1990–2009
(Percent annual change, decadal averages)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: CDSI; and IMF staff calculations.Note: Decomposition based on formula in previous paragraph.Decomposition of Labor Productivity, 1990–2009
(Percent annual change, decadal averages)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: CDSI; and IMF staff calculations.Note: Decomposition based on formula in previous paragraph.Decomposition of Labor Productivity, 1990–2009
(Percent annual change, decadal averages)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: CDSI; and IMF staff calculations.Note: Decomposition based on formula in previous paragraph.28. The results highlight the importance of not only achieving higher productivity in individual sectors but also of moving employment into high-productivity activity. To maximize overall income generation, employment should flow toward the areas where it is most productive. Indeed, Rodrik and McMillan (2011) argue that labor flows from low-productivity activities to high-productivity areas are a key driver of development and explain the bulk of the better productivity performance of Asia compared to Latin America. Although Saudi Arabia’s non-oil manufacturing sector has seen rapid increases in labor productivity, the lack of commensurate increases in employment in this sector has limited the positive impact on economy-wide labor productivity.10
29. It will be important to continue the diversification process and move further up the value chain while also absorbing the growing number of job-seekers. One challenge is that the petrochemical and related industries that represent the fastest-growing part of the economy and account for the bulk of non-oil exports require heavy investment and—like oil production—offer limited potential for large-scale job creation. Another challenge is that this sector is highly energy-intensive and is using an increasing share of the country’s energy production at prices that are substantially lower than those available in foreign markets. To generate sustainable broad-based growth in income and employment it will therefore be necessary to expand production in less energy-intensive and more labor-intensive but still high-paying areas. That in turn, depends on enhancements in the efficiency of factor markets to provide for continued increases in productivity.
30. Continued diversification could potentially add significantly to overall income growth. Over the past three decades, the two countries that have seen the largest increase in the share of non-oil exports in total exports are Indonesia and Mexico—where the share in both cases increased from an average of about 40 percent in the 1980s to more than 80 percent in the 2000s (Figure I.11).11 In other words, they went from levels of dependence on oil for foreign currency earnings that were not that far from where Saudi Arabia is today to a situation where oil represents a minor share of export receipts. This was achieved while maintaining substantially higher growth in overall GDP per capita than in Saudi Arabia—in Indonesia’s case, despite a gradual decline in oil output. The experience of Indonesia and Mexico thus suggests that rapid diversification is not only possible but could also be associated with a large increase in income growth in Saudi Arabia, not least because abundant hydrocarbon reserves put the country in a good position to maintain oil production at or above current levels for many years to come.
Diversification and Growth, 1980–2009
(Percentage points)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: IMF World Economic Outlook database.1 Based on decadal averages.Diversification and Growth, 1980–2009
(Percentage points)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: IMF World Economic Outlook database.1 Based on decadal averages.Diversification and Growth, 1980–2009
(Percentage points)
Citation: IMF Staff Country Reports 2012, 272; 10.5089/9781475510621.002.A001
Source: IMF World Economic Outlook database.1 Based on decadal averages.References
Al-Khatib, M.A., 2011, “The Impact of Economic Diversification on the Saudi Non-oil Economy [in Arabic],” Arab Journal of Administrative Sciences, Vol 18 (2).
Baier, S.L, G.P. Dwyer, and R. Tamura, 2006, “How Important are Capital and Total Factor Productivity for Economic Growth?” Economic Inquiry, Vol. 44 (1), pp. 23–49.
Barro, R. and J. W. Lee, 2010, “A New Data Set of Educational Attainment in the World, 1950–2010,” NBER Working Paper No. 15902, (Cambridge, MA: National Bureau for Economic Research).
Bisat, A. M. A. El-Erian, and T. Helbling, 1997, “Growth, Investment, and Saving in the Arab Economies,” IMF Working Paper No. 97/85, (Washington: International Monetary Fund)
Caselli, F., 2005, “Accounting for Cross-Country Income Differences,” in P. Aghion and S. Durlauf (eds.), Handbook of Economic Growth, (New York: North-Holland).
Chen, V., A. Gupta, A. Therrien, G. Levanon, and B. v. Ark, 2010, “Recent Productivity Developments in the World Economy: An Overview from The Conference Board Total Economy Database,” International Productivity Monitor, Vol. 19, pp. 3–19.
Denison, E. F., 1962, The Sources of Economic Growth in the United States and the Alternatives Before Us, (New York: Committee for Economic Development).
Everhart, S. and R. Duval-Hernandez, 2001. “Management of Oil Windfalls in Mexico: Historical Experience and Policy Options for the Future.” Policy Research Working Paper WPS 2592, (Washington: World Bank).
Gala, P., 2007, “Real Exchange Rate Levels and Economic Development: Theoretical Analysis and Econometric Evidence,” Cambridge Journal of Economics, Vol. 32, pp 273–288.
International Monetary Fund, 2011, Qatar: 2010 Article IV Consultation: Staff Report, Statement by Staff Representative, and Public Information Notice Following Consultation, IMF Staff Country Report No. 11/64 (Washington)
Keller, J. and M. K. Nabli, 2002, “The Macroeconomics of Labor Market Outcomes in MENA over the 1990s: How Growth has Failed to Keep Pace with a Burgeoning Labor Market,” The Egyptian Center for Economic Studies Working Paper No. 71, (Cairo: The Egyptian Center for Economic Studies).
Palma, G. 2004, “Four Sources of De-industrialization and a New Concept of the Dutch Disease,” in J. A. Ocampo (Ed.), Beyond Reforms: Structural Dynamics and Macroeconomic Vulnerability, (Palo Alto, CA: Stanford University Press).
Rodrik, D. and M. S. McMillan, 2011, “Globalization, Structural Change, and Productivity Growth,” NBER Working Paper No. 17142, (Cambridge, MA: National Bureau of Economic Research).
Sala-i-Martin, X. and E. Artadi, 2002, “Economic Growth and Investment in the Arab World,” in K. Schwab and P. Cornelius (Eds.), The Arab World Competitiveness Report 2002–03, (New York: Oxford University Press).
Solow, R. M., 1957, “Technical Change and the Aggregate Production Function,” Review of Economics and Statistics, Vol. 39 (3), pp. 312–20.
Prepared by Ghada Fayad and Tobias Rasmussen.
The growth accounting methodology was pioneered by Solow (1957) and further developed to distinguish between labor quantity and quality by Denison (1962).
See Chen et al. (2010) and Conference Board Total Economy Database, January 2012, available at http://www.conference-board.org/data/economydatabase/.
TED divides physical capital into two groups, non-ICT capital and ICT capital. We omit this decomposition, as data for ICT capital are not available for Saudi Arabia.
The labor cost share in the United States and other advanced economies is typically found to be about 0.65. Where direct measurements are not available, many studies simply apply a constant value for the labor cost share of about that magnitude. The lower value used in TED is motivated by the observation that capital in emerging and developing economies tends to be scarce and thus earns a relatively high rate of return, while abundant labor suppresses wages and implies a lower labor cost share. For Saudi Arabia, an additional complication is that the highly segmented labor market, with limited direct competition between Saudi and non-Saudi workers, raises questions about the applicability of standard labor cost share values.
For Saudi Arabia and many other countries, data for skill categories in TED are adapted from estimates of educational attainment from Barro and Lee (2010), with low-skill labor defined as no or only primary education, medium-skill as secondary education, and high-skill as tertiary education. The Barro-Lee data for Saudi Arabia are shown in Figure I.1.
This assumes an initial steady-state relationship. For Saudi Arabia, our constant price investment series start in 1969 and we use δ = 0.06 (a commonly applied figure in the literature) and g = 0.05 (near the average for emerging markets in the 1960s).
The result that overall TFP growth in Saudi Arabia has been low is robust to alternative parameterization. For example, with a labor cost share of 0.7 rather than 0.5, the average annual rate of TFP growth during 1990–2009 changes by less than 0.1 percent. The same is true for a doubling of the depreciation rate, δ, to 0.12, or of the initial growth rate, g, to 0.10. Sensitivity to the construction of the human capital stock is somewhat greater than for the physical capital stock. An alternative specification for calculating labor quality that uses a piecewise linear function of years of schooling, as in Caselli (2004), reduces TFP growth to -0.3 percent (0.1 percent for non-oil GDP). TED’s estimate of TFP growth in Saudi Arabia is -0.6, with the difference from the present estimate largely due to their series for raw labor (which is based on working-age population) showing higher growth than the series applied here (which is based on actual employment growth).
The prevalence of low rates of TFP growth across countries is a well-established result (see e.g. Baier et al., 2006) but should not be taken to mean that TFP growth is unimportant. Indeed, development accounting—a variant of the growth accounting literature that looks at cross-section rather than cross-time differences—shows that the bulk of variation in income across countries can be attributed to differences in TFP (Caselli, 2005).
This is consistent with Rodrik and McMillan (2011), who find that labor movements in countries with a relatively high share of natural resources in exports have been growth-reducing.
Among the many contributing factors to Mexico and Indonesia’s successes toward economic diversification, some elements have been highlighted in the literature. These include Mexico’s management of inherently volatile oil revenues with a stabilization fund working as the main recipient of revenues, and the use of financial markets to hedge short-lived movements in prices (Everhart and Duval-Hernandez, 2001). Indonesia’s success, like that of other East Asian countries, is often attributed to an export-led growth strategy along with high investment and savings levels (Gala, 2007, and Palma, 2004).
