Journal of Asian Economics 14 (2003) 735–757

The growth–equity trade-off in modern economic development: the case of Thailand$ Medhi Krongkaewa,*, Nanak Kakwanib a

School of Development Economics, National Institute of Development Administration (NIDA), 118 Serithai Road, Bangkok 10240, Thailand b School of Economics, University of New South Wales, Sydney, NSW, Australia

Abstract The trade-off between economic growth and income inequality is quite clear in the case of Thailand during the last four decades of its development. Rapid economic growth had brought about rapid reduction in poverty while income inequality had risen. The paper also shows that when poverty reduction is decomposed into two separate effects of growth and income distribution on poverty reduction, the growth impact on poverty reduction was lessened by the incidence of high income inequality. In short, the growth in Thailand was not a pro-poor growth. The high income inequality can be explained, in part, by unequal returns to productive inputs in imperfect markets and unfair competition, by unequal landownership and unsuccessful land reforms, and by political and administrative structures that protect the positions and interests of the relatively well-off. To help reduce the growth–equity trade-off, future growth in Thailand must be pro-poor or more pro-poor. # 2003 Elsevier Inc. All rights reserved. Keywords: Thailand; Growth; Income distribution

1. Introduction The trade-off between growth and equity in economic development is as old as economics itself. Economics, as a social science, strives to explain human behaviour in making choices between or among sets of alternatives, each of which has associated with it some notions of opportunity costs. What this means is that there is a cost, and/or benefit, associated with every human decision. It is not unusual, therefore, to recognise that development decision or policy that brings about growth may also bring about an $ Earlier version of this paper was presented at the session organised by the JBIC Institute at the Fourth Global Development Network (GDN) Conference in Cairo, 19–21 January 2003. * Corresponding author. Tel.: þ66-2-727-3196; fax: þ66-2-375-8842.

1049-0078/$ – see front matter # 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.asieco.2003.10.003

736

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

alternative cost of growth, of which increased income inequality is a major economic phenomenon. It is possible that this trade-off does not occur, that is to say, growth does not necessarily bring about increased inequality but, on the contrary, income equality may improve with growth. In the past 30 years, economic literature is replete with arguments for and against the existence of this trade-off,1 but it seems that, at least in East Asia, such an argument has lost its importance during the high-growth periods of the 1980s and 1990s because as long as the economy continues to grow rapidly, that everyone’s pocketbook or wallet is bulging, few would be concerned with the issues of income inequality. But growth without equality or growth with high income inequality is not a sustainable growth. The recent economic crisis in East Asia is a good indication of such vulnerability. Indeed it could be argued that the crisis that started first in Thailand in 1997 was brought about by the deep-rooted inequality situation in the Thai economy and society. The main purpose of this paper is to address the issue of the trade-off between growth and equity in Thailand as a typical, rapidly-growing modern economy in East Asia. The use of Thailand of a case study is important in a sense that, while the economic development of Thailand exhibits patterns and traits that can also be seen in other East and South-East Asian countries such as Korea and Malaysia, there is also uniqueness associated with Thailand as the only country in South-East Asia never colonised by Western power during the 19th century colonisation push, and the only country in South-East Asia that the overseas Chinese have been completely assimilated with the indigenous Thais during the last 40 years of economic development. This mixture of common or universal and specific characteristics that determine, and/or are influenced by the growth–equity nexus can be used to explain a unique character of Thailand’s modern economic development experience. Normally when a country embarks on its journey to economic development, it faces a series of questions concerning that path to development. One is the question on what development procedure or philosophy to follow: capitalistic or socialistic or a mix of the two. Then other questions may include: Who is to decide on such procedure? What would be or should be the speed and tempo of such development? Who would gain and lose or how the benefits and costs of development are shared among the people in the country? and so on. In a democratic political system, the first question will need to be answered politically through choices of politicians or political parties representing certain economic and political ideas. The process may be long drawn out as everyone is involved in the choice or selection process. But for Thailand when it embarked on its development journey in the early 1960s, this was not a difficult question at all, as the country was run by a military dictator who decided the path of development on behalf of the whole population. Field Marshall Sarit Thanarat was that military dictator when the First National Economic Development Plan was announced in 1961, heralding the start of Thailand’s modern economic development. Lest the above event be looked upon only in bad lights, it should be mentioned at the outset that many have regarded Field Marshall Sarit as the Benevolent Dictator who ushered in the new era of development to the Thai people.2 He had made a great use of 1

See, for example, Okun (1975). For a classic treatment on the life of Field Marshall Sarit and how he led Thailand in the late 1950s and early 1960s, see Charoemtiarana (1979). 2

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

737

advice from astute technocrats during his time, and put in place many rules, regulations and institutions that, undeniably, helped Thailand grow very fast after the launch of its First Plan. On looking back, his doing, and undoing, could be a part of the reasons that can be used to explain the apparent trade-off between rapid growth and increasingly unequal distribution of income of the Thai people as the country grew. But perhaps the real reasons went much deeper and farther back. There is no one single reason that explains the Thai growth–equity trade-off. The reasons are many, as we shall see. Once the question ‘‘Who determined economic development policies?’’ is solved, the question ‘‘What kind of development policies to pursue?’’ became critical. The late 1950s were the periods of the Cold War between the US and the Soviet Union and China. Many countries around Thailand had succumbed to communist influences, leaving Thailand the ‘front line’ state that the US needed to protect and keep on its side of the Free World. Therefore, the US government poured in military, economic and technical assistance to keep Thailand strong militarily and economically. A part of the US economic help was through the technical assistance of the World Bank which sent in a team of economic experts before Sarit’s time, but the report of this World Bank Mission Team has become the ‘Development Bible’ that guided Thailand in the first one or two decades of its development planning.3 It may look as though Thailand had no choice but to go along with the free-market, capitalistic direction provided by the US, but the truth was that this was also what Sarit himself would like to see Thailand develop. Thus a special relationship between Thailand and the US had developed and last until the end of the Vietnam War in the last 1970s.4 Sarit was dead in 1964, half way through his First 6-Year Plan, but the momentum of growth was already established. His successor, another military prime minister, continued on Sarit’s development legacy until 1973 when he was toppled by a popular, student-led, uprising. By that time (during the Third Plan) it was obvious that rapid growth of the economy had brought about increasing income inequality. The change of Thai government and regime in 1973 had spurred a flurry of economic activities that aimed at social equality and public welfare issues. The short-lived return to democracy in 1974 was ended in 1976 with the fall of democratically-elected government by another military seizure of power. From then on the country had gone into a series of economic and political changes that have direct bearings upon the overall growth and equality in Thailand until today. The above account is just a glimpse of what happened to growth and equity situations in Thailand in the first 10 years or so of its modern economic development. In what follows, 3 This World Bank (1959) report was later published as a book titled: A Public Development Program for Thailand. For an excellent account on the relationship between Thailand and the US in the 1950s to the 1980s, see Muscat (1990). Muscat (1966) also wrote an earlier book on the early period of development planning in Thailand. See his ‘‘Development Strategy in Thailand’’. 4 Two political events in East Asia that occurred in 1962 that help judge the success or failure of Thailand development policies were the seizure of power in Burma by General Ne Win and in (South) Korea by General Park Jung Hee. Burma under Ne Win had turned inward into ‘Burmese Ways to Socialism’, whereas Korea under Park had adopted the ‘Development State’ model where the government provided political and economic support to the private sector who would be the main actor in the overall development efforts. The differences between Burma and Korea after 40 years of these different development policies were obvious. Burma practically stopped growing, whereas Korea has succeeded in becoming a newly developed economy and was able to join the OECD in 1995. Thailand’s growth patterns are similar to Korea’s, although it is often said that Korea was about 10–15 years ahead of Thailand in terms of economic development.

738

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

Table 1 Sectoral share of GDP from 1960 to 2000 (%)

Agriculture Industries Services GDP

1960

1970

1980

1990

1995

2000

31.5 19.7 48.8

27.3 24.0 48.7

20.2 30.1 49.7

13.6 37.8 48.6

10.8 41.7 47.5

11.4 43.1 45.5

100.0

100.0

100.0

100.0

100.0

100.0

Source: Adapted from Tinakorn (2002).

we will analyse growth and equity in the Thai economic development in more details. The next section (Section 2), we will look at records of growth, poverty and income distribution of the Thai population in the past four decades. In Section 3, we will explain how we measure the separate effects of growth and equity on poverty, leading to Section 4 where we will try to explain the reasons for such growth–equity trade-off. Section 5 summaries and discusses some important policy implications.

2. Records of growth and equity in the Thai economic development When the first National Development Plan was launched in 1961, Thailand was a typical agricultural economy where over 80% of the population were engaged in agricultural activities with rice as a major crop for both domestic consumption and export. Other major crops and primary products for export included rubber, maize, kenaf, teak and tin. The rate of growth of the economy was comfortable, albeit low, at about 2–3% a year. Under the threat of population explosion (the rate of growth of population was over 3% a year), the government decided to move the economy away from a low-level equilibrium, and push it into an era of systematic development. The basic development philosophy behind the 1961–1966 National Economic Development Plan was that the government would provide necessary infrastructure for the economy and reorganise government agencies to facilitate orderly economic transformation, but otherwise the private sector would be encouraged to engage in productive investment and transactions.5 Table 1 shows the change in the structure of the Thai economy from the First Plan in 1960–2000, corresponding to the implementation of the First National Economic Development Plan to the Eight National Economic and Social Development Plan. In 1960, just before the launch of the First Plan, the GDP share of agriculture was 31.5%, whereas the share of industries and services were only 19.7 and 48.8%, respectively. From this First Plan period onward, the relative importance of agriculture began to decline, slowly at first, but accelerated when the industrial and manufacturing sectors became more established in 5 The relevant wording in the plan is that the plan was of the view that in Thailand ‘‘. . . increased output will be most readily secured through the spontaneous efforts of individual citizens, fostered and assisted by Government other than through Government itself entering into the field of production’’, and also, ‘‘. . . the key note of the public development is, therefore, the encouragement of economic growth in the private sector and the resources of Government will be mainly directed to projects, both in the agricultural and nonagricultural sectors of the economy, which have this objective in view’’.

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

739

Table 2 Sectoral share of employment from 1960 to 2000 (%)

Agriculture Industries Services GDP

1960

1970

1980

1990

1995

2000

82.3 4.2 13.5

79.3 5.9 14.9

70.8 10.3 18.9

64.0 14.0 22.0

52.0 19.7 28.2

48.8 19.0 32.2

100.0

100.0

100.0

100.0

100.0

100.0

Note. The figures from 1960 and 1970 are from the Population Census, whereas those figures for 1980–2000 are from the Labour Force Surveys. Source: Adapted from Tinakorn (2002).

the 1980s and 1990s. In 2000, for example, the GDP share of agriculture was 11.4%, whereas that of industries and services were 43.1 and 45.5%, respectively. The share of the industrial sector exceeded that of the agricultural sector around 1984, signifying a fulfilment of conditions for a newly industrialised country. But Thailand was far from industrialised country if one looks at its employment pattern. As shown in Table 2, in 1960, about 82.3% of the Thai population were engaged in agriculture, whereas on 4.2 and 13.5% were engaged in industries and services, respectively. In 2000, still almost half of the Thai population (48.8%) were found in its agricultural sector, whereas 19.0 and 32.2% were found in industrial and services sectors, respectively. This alone is sufficient to illustrate the inequity between the agricultural and industrial and services sectors.6 The Thai economy has arrived at where it is today as a result of the combination of the efforts of the public and private sectors. As mentioned earlier, the development philosophy of the Thai government was to depend on private initiatives for investment and production, with the government providing infrastructure and rules an regulations based mainly on free enterprise system. The government did intervene in the market through investment promotions and industrial protection, but not to the extent that the role of the Thai government could be identified as a ‘developmental state’ where ‘picking the (industrial) winners’ is the norm rather than the exception. Yet the orientation and attention towards the industrial sector as against the agricultural sector was unmistakable. The ‘exploitation’ of the Thai agricultural sector in the early phases of economic development, especially in the periods up to the early 1980s, was well documented and explained.7 In other words, Thai economic development owes a lot to its agricultural sector in the early phases of its development. Table 3 sums up the growth rates of each of the three sectors and overall GDP from 1960 to 2000. On the whole, the growth of the Thai economy was among the most rapid in the world. For example, during the First Decade of the 1960s, the average growth rate of its 6 When 48.8% of the employed people in agriculture contributes only 11.4% of GDP, while 19% of employed people in the industrial sector contributes 43.1%, it is obvious that the productive return of workers in the industrial sector is much greater than that of the workers in the agricultural sector. A simple productivity ratio between GDP and employment share in the agricultural sector was only 0.23 while that of the industrial sector was 2.27, almost a 10-time difference. 7 See, for example, Siamwalla and Setboonsarng (1989) and World Bank (1959).

740

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

Table 3 Average growth rates of GDP of Thailand from 1961 to 2001 (%)

Agriculture Industries Services GDP

1961–1970

1971–1980

1981–1990

1991–1995

1996–2000

1961–2000

6.4 10.0 7.8

3.6 9.2 6.9

3.7 10.4 7.7

3.9 10.8 8.1

1.4 1.4 0.3

4.1 9.0 6.6

7.8

6.7

7.9

8.7

0.6

6.8

Source: Adapted from Tinakorn (2002).

GDP in real term was 7.8%. This had reduced to 6.7% during the Second Decade of the 1970s, but gone back to 7.9% during the 1980s. This growth in the Third Decade reflected a mixed picture of economic situations within the decade. During the first half the 1980s, Thailand faced the negative impacts of the second oil shock resulting its external imbalance and the need to seek the first balance of payments assistance from the IMF. But the second half of the 1980s saw a complete change in the economic situations. The significant devaluation of the baht in 1984, the revaluation of the yen after the Plaza Accord in 1985 which resulted in massive inflows of foreign direct investment into Thailand from Japan, and the success in the large campaign of the Thai tourist industries which started in 1987 had pushed the growth of the Thai economy onto another, higher growth path. The last 3 years of the 1980s namely 1988, 1989 and 1990, the growth of GDP of Thailand was all double-digit, that is 13.3, 12.1 and 10.0, respectively. This was the beginning of the bubble economy which last about 10 years before it began to burst in mid-1990s. The growth slowed during the first few years of the 1990s due to the effects of the Gulf War and the internal political turmoil in the country but still very high. Indeed, the average growth rate during the first 5 years of the 1990s was 8.7%. But this growth figure belies the vulnerability of the Thai economy where the inflows of unprotected short-term capital eventually led to the collapse of exchange rate regime and the loss of almost all of the 38 billion US dollars foreign reserves through the government’s futile defence against foreign and local exchange rate speculators.8 The rapid outflows of the above short-term capital was a major cause of currency crisis that had turned into financial crisis, and eventually economic crisis that engulfed not only Thailand but many parts of East Asia and indeed the world as well. The growth rate of GDP became negative for the first time in its recorded history in 1997, and was as low as 10.5% in 1998, before recovering to 4.4% in 1999 and 4.6% in 2000. The overall growth rate of GDP of Thailand from 1961 to 2000 averages about 6.8% per annum in real term. Another way of looking at the growth of the Thai economy is to use the results of household income surveys. The aggregate income numbers from these surveys are smaller than national income figures because they include only income of households and not that of the government and corporate sectors. But perhaps this household income is a better 8 There are several accounts of the causes and development of the Thai economic crisis by Thai economists. See, for example, Siamwalla (2003), Medhi Krongkaew, ‘‘A Tale of an Economic Crisis’’, in Chu & Hill (2001).

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

741

Table 4 Household income per capita per month by region (current prices) North Income (baht per month) 1962/1963 90 1968/1969 153 1975/1976 307 1981 704 1986 796 1988 930 1990 1336 1992 1651 1994 1825 1996 2450 1998 2822 2000 2538 Annual growth rate (%) 1962/1963 1968/1969 1975/1976 1981 1986 1988 1990 1992 1994 1996 1998 2000

North-East

Central

South

BMR

WK

83 132 253 493 521 650 909 1225 1418 1847 2098 1990

98 233 433 852 954 1062 1566 2143 2460 3030 3509 3715

152 171 337 740 871 936 1406 1920 2050 2525 2891 2897

196 333 604 1422 1829 2342 3406 4982 5390 6858 7419 7794

133 208 351 751 844 1064 1555 2130 2321 2913 3342 3358

9.2 10.5 14.8 2.5 8.1 19.9 11.2 5.1 15.9 7.3 5.2

8.0 9.7 11.8 1.1 11.7 18.3 16.1 7.6 14.1 6.6 2.6

15.5 9.3 11.9 2.3 5.5 21.4 17.0 7.1 11.0 7.6 2.9

2.0 10.2 14.0 3.3 3.7 22.6 16.9 3.3 11.0 7.0 0.1

9.2 8.9 15.3 5.2 13.2 20.6 20.9 4.0 12.8 4.0 2.5

7.7 7.8 13.5 2.4 12.3 20.9 17.0 4.4 12.0 7.1 0.2

Memo item: average growth rate 1962/1963 to 2000 9.2

8.7

10.0

8.1

10.2

8.9

Source: NSO (various years).

indication of economic welfare as it represent income of the people rather than that of agencies or companies. The National Statistical Office (NSO, 2003) of Thailand has been conducting household income and expenditure surveys since 1962, and their statistical data have formed a critical basis for the computation of poverty incidence and the distribution of income of the Thai people since 1962. Table 4 shows the average monthly per capita household income of the Thai people classified by different regions. From Table 4, it may be seen that the average per capita household income had increased from 133 baht in 1962/1963 to 3358 baht in 2000. The overall average increase of this household per capita income was estimated at 8.9%. This is even higher than the average growth rate of GDP from 1961 to 2000. If we can accept that the growth in household income from these household income surveys is uniform with the growth in national income from national income accounting, then income of the household sector grows faster than the overall national income. Across the five regions of Thailand, the growth of income of the population of Bangkok and its vicinity was faster than the rest of the country at

742

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757 9,000 8,000

N.

NE

C.

7,000

S.

BMR

WK.

Current P

6,000 5,000 4,000 3,000 2,000 1,000 0 1962/63 1968/69 1975/76

1981

1986

1988

1990

1992

1994 1996new1998new 2000new

Plate 1. Household income per capita per month by region.

10.2% between 1961 and 2000, followed by that of the Centre, the North, the North-East, and the South at 10.0, 9.2, 8.7 and 8.1%, respectively. It may be noticed that the average growth rates in all regions in 40 years of economic development did not differ much from one another. What this means was that the growth in Thailand seems to spread evenly throughout all regions. This may sound good, but what was not good about this was that if the interregional income differences are large from beginning, then 40 years of economic development did little to narrow these interregional income differentials. This was exactly the case when we look at the average income of the people in each region in the upper part of Table 4. Income of the North-Eastern households was always the lowest, and that of Greater Bangkok was always the highest. The ranking of each region has never changed throughout these four decades, but the average income of the Bangkok population continued to diverge from the rest of the country, as Plate 1 amply demonstrates. From this chart, it may be seen that in the year 2000, the average income of Bangkok household was more than twice the national average, and more than three times the average of the North-East. The lop-sided prominence of Bangkok could be one of the reasons that causes the growth disparities which brings about greater income disparities, as we shall see presently. Growth in average income across the board must necessarily cause poverty incidence to fall. As shown in Table 5, the incidence of poverty estimated from the first household income and expenditure survey in 1962/1963 was a high as 57% of the total household population.9 This poverty incident successively reduced as the country grew. As there are 9 See Medhi Krongkaew, ‘‘Changes in Poverty and Income Inequality in Thailand, 1962/1963 to 1992’’, Working Paper Series-WP1/96, Economic Growth, Poverty and Income Inequality in the Asia-Pacific Region Project, School of Economics, University of New South Wales and Faculty of Economics, Thammasat University, 1996.

1962/1963

1968/1969

Poverty incidence head-count ratio Series I 57.0 42.0 Series II Series III Series IV Income inequality (a) Gini ratio Series A Series B Series C

0.563

0.555

1975/1976

1981

33.0 30.0

31.3 23.0

0.605 0.426

(b) Quintile share Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Total (c) T/B ratio Source: Computed from various Socio-Economic Surveys.

0.453

1988

1990

1992

1994

1996

1998

2000

21.2 22.8 32.6

18.6 27.2

13.7 23.3

16.3

11.4

13.0

16.2

0.479 0.4508

0.504 0.4809

0.543 0.4989

0.4861

0.4775

0.486

0.4901

5.23 9.00 13.44 21.13 51.20

4.93 8.42 12.65 19.91 54.10

4.53 7.94 12.17 19.75 55.62

4.62 8.29 12.74 20.27 54.07

4.77 8.46 12.92 20.50 53.34

4.63 8.30 12.63 20.20 54.23

4.47 8.14 12.50 20.33 54.56

100.00

100.00

100.00

100.00

100.00

100.00

100.00

9.8

11.0

12.3

11.7

11.2

11.7

12.2

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

Table 5 Incidence of poverty and income inequality

743

744

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

several series of poverty incidence estimation, especially in the early periods of poverty studies, these poverty incidences may not strictly be compatible, but the trend in poverty reduction was not in doubt.10 From 1988 onward where the newest poverty lines were created and used to identify the poor across Thailand, about a third of the total Thai population (32.6%) could be classified as poor in 1988. Recall that the growth rate of GDP was highest in 1988. This should translate into some positive effects on poverty reduction. This was so as the incidence of poverty fell drastically between 1988, 1990 and 1992. Based on this newest series of poverty estimates, the reduction in poverty incidence as almost 10% points between 1988 and 1992, or the annual rate of poverty reduction of about 8.1%. This poverty reduction continued until 1996 when the incident was 11.4%. Then the crisis hit in 1997, resulting in the slight increase in overall poverty incidence of 13.0% in 1998. The full impact on poverty was felt in 2000 when the incidence had reached 16.2%. The secular trend in poverty reduction in Thailand has been broken. Another trend was also broken in the last few years. This was the trend of the improvement in income distribution in Thailand since 1992. As shown in Table 5, the distribution of income of the Thai people started to become more unequal from 1962/1963 onward, that is as soon as the country announced and implemented its First National Economic Development Plan. Despite different estimates of income inequalities, the increasing trends of income inequality were also unmistakable. Again from 1988 where the new income distribution series was used to compute poverty incidence, the new series was also used to compute income inequality. The inequality as measured by the Gini ratio or Gini coefficient was 0.4508 in 1988, increasing to 0.4809 in 1990 and 0.4989 in 1992. Then in 1994 the Gini ratio began to fall. Referring to the famous Kuznets curve where the index of income inequality on the vertical axis is shown to rise with the increase in the national income on the horizontal axis in the early part of economic development until a certain level of development is reached when this index of inequality begins to decline, the income inequality in Thailand appeared to have reached the apex of this inverted U-shape Kuznets curve in 1992, and this increasing trend of income inequality turned downward from 1994 onward. However, this improvement in income inequality had proven to be short-lived. In 1998, at the depth of economic crisis, the income inequality increased again. The Gini coefficients for 1998 and 2000 were 0.4860 and 0.4901, respectively. In other measures of income inequality, the quintile shares of income of the Thai people in Table 5 also shows the income share of the poorest, bottom quintile (Quintile 1) reaching the lowest point of 4.53% in 1992 before increasing to 4.62% in 1994. At the same time, the income share of the top or richest quintile (Quintile 5) had reached 55.62 in 1992 before falling to 53.34% in 1994. In terms of the Top to Bottom Quintile Ratio or T/B ratio, this ratio was 12.3 in 1992, falling to 11.7 in 1994 but reverted to 11.7 again in 1998 and further to 12.2 in 2000. The income inequality situation in time of crisis was still in state of flux. Before considering the reasons for the trade-off between growth and inequality in the Thai case, we want to ascertain whether the growth–equity trend in Thailand really follows the Kuznets hypothesis. Dr. Pranee Tinakorn (2002) of Thammasat University has already provided some answers to this question. Her quadratic regression of Gini ratios on constant 10

For details on the computation of each poverty incidence series, see Krongkaew (1996).

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

745

GINI .50

.49

.48

.47

.46 Observed .45

Quadratic

1000

2000

3000

4000

Plate 2. Relationship between household income and index of income inequality. Note. Gini ¼ a þ b lnðYÞþ c lnðYÞ  2, where Y is household income. Data from Tables 4 and 5.

GNP of Thailand from 1975/1976 to 2000 had shown a good fit of these two critical variables.11 We have also tried to fit our Gini coefficients (in Table 5) to our household per capita income (in Table 4) from 1988 to 2000. The result which was shown in Plate 2 also indicates a close relationship between economic growth and income inequality. As the rising trend of income inequality during 1998 and 2000 was small, this may have little long-term impact on the long-term trend of income distribution. From Plate 2 it may be seen that the quadratic trend line of the income distribution in Thailand in 2000 is still sloping downward. How believable or reliable is this conjecture that income inequality in Thailand is likely to fall as the economy recovers from the crisis? We would like to argue that income inequality has much greater impacts on poverty reduction in Thailand than many have previously thought. While it is obvious that economic growth which manifests itself through an increase in average income in all groups of the population across all regions has helped bring about a reduction in poverty, what was not obvious was how this poverty reduction would be different with different 11 Dr. Pranee regressed Gini coefficients of Thailand from 1975/1976 to 2000 on GNP per capita on constant 1988 price for the same periods. The result obtained was:

Gini ¼ 24:4845 þ 0:001391Y  0:0000000172Y 2 ; ð9:224Þ

ð8:194Þ

DurbinWatson statistic ¼ 2:323;

ð6:933Þ

R2 ¼ 0:9117;

adjusted R2 ¼ 0:8588;

F-statistic ¼ 17:216:

Observations: SES 1975/1976 to SES 2000; value in parenthesis is t-statistic, where Gini: Gini coefficient (with base adjusted to 100), and Y: GNP per capita in constant 1988 price.

746

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

levels of income inequality. On this point, we would exert that economic growth in Thailand in the last 30 years since the beginning of the First Plan in 1961 had created too much of a trade-off on income distribution. The worsening of this income distribution had slowed the speed of poverty reduction. In other words, poverty reduction in Thailand would have been much faster had the income distribution in Thailand been more equal. In the next section we will demonstrate how we measure the separate influences of growth and equity on poverty reduction.

3. The growth–equity trade-off: how to measure it and how to interpret the results The notion that growth is good for the poor is well accepted, and is said to be influenced by a well known study by Dollar and Kraay (2000). However, as Kakwani (2000 and 2002) and Kakwani and Pernia (2000) have pointed out, the different nature of growth could have different effects on poverty. For some countries, the growth-maximising policies alone may be adequate to provide satisfactory reduction to poverty, but for many countries there may be a need to have pro-poor growth policies with a focus on reducing inequality before or concurrently with a focus on increasing growth. This last notion is based on the belief that the degree of poverty reduction depends on both income and its distribution. The increase in average income reduces poverty and the increase in inequality increases it. Therefore, the change in poverty can be decomposed into two components: one is the growth component relating to change in mean income, and the other is the inequality component relating to change in inequality. Kakwani and Pernia (2000) have developed an index of pro-poor growth, which is based on a decomposition of total change in poverty into (a) the impact of growth when the distribution of income does not change, and (b) the impact of income redistribution when total income does not change. Suppose Z is the poverty elasticity with respect to growth, which is defined as the proportional change in poverty when there is a positive growth rate of 1%, then Z can be decomposed into two components, Zg and ZI such that Z ¼ Zg þ Z I

(1)

where Zg is the pure growth effect and ZI is the inequality effect. Zg is the proportional change in poverty when the distribution of income does not change, whereas ZI is the proportional change in poverty when inequality changes in the absence of growth. Zg will always be negative, meaning that a positive growth always leads to poverty reduction, with distribution remaining constant. ZI can be either negative or positive depending on whether change in inequality accompanying growth reduces or increases poverty. Growth will obviously be pro-poor if ZI is negative. Thus the degree of pro-poor growth can be measured by an index Z (2) f¼ Zg f will be greater than 1 when ZI < 0. Growth will be pro-poor when f > 1, meaning that the poor benefit proportionally more than the non-poor, that is to say growth has resulted in

M. Krongkaew, N. Kakwani / Journal of Asian Economics 14 (2003) 735–757

747

a redistribution in favour of the poor. This would be the first-best outcome. When 0 < f < 1, growth is not strictly pro-poor (i.e. growth results in a redistribution against the poor) even though it still reduces poverty incidence. This situation may be generally characterised as ‘trickle-down’ growth. If f < 0, economic growth actually leads to an increase in poverty. This is equivalent to what Bhagwati would call ‘immiserising’ growth. Index f measures how the benefits of growth are distributed across the population. Suppose g is the growth rate and y is a poverty measure, the proportional change in poverty may be written as Dy ¼ f ðg; fÞ ¼ f ðg ; 1Þ (3) y where @f ðg; fÞ @f ðg; fÞ < 0 and