We tend to associate migration with poverty. It reminds people that, for example, poor people in rural areas who cannot earn a livelihood in their hometown are forced to migrate to urban areas to earn money and send back a part of the money to their parents. However, migration is not always pessimistic. It, at the same time, means a chance for younger people to succeed. There seems to be a turning point at which the perception of migration changes from filial duty to economic opportunity. However, it is not easy to show when the turning point exactly is, as demonstrated by the various images of migration that tend to obscure the nature of the problem. There are several reasons why we still look at migration with pity. Researchers tend to emphasize the misery of migration because they need to show that their topic is a serious and important problem in developing countries. Politicians vying for bigger central government budgets will also exaggerate the situation, insisting that their constituency is very poor or hardest hit by economic crises when it happens. Non-governmental Organizations (NGO)'s that aim to alleviate poverty are looking for poverty as a chance to do something for “poor” people, or as a good reason to plan their projects, even if there is no severe poverty any more. Official Development Assistance (ODA) organizations are also looking for “poor” people for the same reason. On the other hand, people who know the benefits of the projects of these organizations are willing to be identified as “poor.” Thus the images of poverty are increasingly politically motivated. This, in turn, has the ironic effect of leaving the real needy neglected.

Since the 1997 economic crisis in Thailand, many reports on its impact have been published, most of which emphasized its negative impacts on the poor.⁴ But if you had a chance to visit villages far away from Bangkok at that time, you would have found that people are doing things as usual, as if nothing had happened. For them the crisis was something to be read in newspapers or watch on TV. In fact, the devaluation of the baht in 1997, from which the crisis started, was a chance for exporters and producers for export because the domestic prices of their products increased due to devaluation. Export companies could expect increase in profit unless they had a big foreign debt.⁵ Export-oriented manufacturing companies were very busy receiving customers from foreign countries who wanted to buy now very cheap products from Thailand.⁶   Export companies of agricultural products also gained from the devaluation by the same reason. Farmers producing for export market were also benefited from the devaluation unless the profits were squeezed by the middlemen. However, TV programs and newspapers reported negative aspects of the crisis with much emphasis. It was immoral to report success stories when some people were desperate, losing everything. Foreign media also did the same. Now Thailand became a “poor” country worthy of compassion and foreign aid.⁷

It is true that many migrants lost jobs during the crisis and returned back to their villages. The idea that “poor” migrants lost jobs and returned to “poor” village implies that they were struggling in extreme poverty even when they returned to their villages. Now Northeast Thailand became the center of attention because it had long been regarded as the poorest region in Thailand and had been sending out the largest number of migrants to Bangkok. A significant number of projects were proposed to alleviate poverty in the region, many of which were very questionable in achieving the alleged purpose. Especially after the end of the period of austerity budget by the Democrat Government in February 2001, a large amount of the budget seemed to be allocated to the Northeast and the North. On the other hand, the South, which is a resource-rich region and enjoyed a higher level of income than the North and the Northeast, is now suffering from falling primary commodity prices but the South is still neglected because of the inertia of the idea that poverty is a phenomenon in the Northeast and perhaps because the region is a constituency of the opposition party.⁸

The organization of this paper is as follows: in the next section we examine the income gap between the Northeast and Bangkok as an incentive for migration to show that the regional gap was decreasing in the 1990s. In Section II we examine the regional gap in greater detail by using the Socio-Economic Survey. Our conclusion is that the income gap is not as great as generally believed by migrants, which means that labor market reached equilibrium.⁹ In Section III we estimate the number of migrants between regions. Our finding is that the one-way migration from the Northeast to Bangkok is only a part of the migration in Thailand. Migrants are moving in various directions and a considerable number of people in Bangkok are returning to the Northeast and other regions. This confirms our primary assertion that the image of poor Northeasters flooding into Bangkok is misleading. Section IV examines how people came to believe that the Northeast was hardest hit by the crisis. The conventional idea that migrants are very poor was reinforced by the efforts of many organizations and researchers to look for poor people. Finally, we discuss in the last section the problems generated by adhering to old and what we have found to be inaccurate images of migration in Thailand.

I. Decreasing Income Gap in the 1990s

In a rapidly developing country like Thailand, migration has various aspects and meanings for the life of people, such as better jobs, higher education, and the enjoyment of urban life. However, the conventional idea that connects migration to poverty is so pervasive that it is very difficult to be free from it. Migration from the Northeast region is also the case.

Actually there are many kinds of statistics that convince us that the Northeast is hopelessly poor. The most extreme figures can be found in the statistics of per capita Gross Regional Products (GRP). Table 1 shows that the per capita GRP of the Northeast is only 12 – 13 percent of Bangkok's, which means that the Northeast earns on average only one eighth of Bangkok. This regional gap may be among the largest in the world. Even though the income gap decreased significantly in 1998, a year after the crisis, the gap began to widen again except for the Central region. If this figure reflects reality, we have to ask how such a huge income gap has gone unchecked. Usually economists believe that a large income gap will be narrowed after a significant number of people migrate to urban areas, as in the dual economy model^.10 However, in spite of several decades of migration, income gap is far from narrowing but still keep widening.

Another statistic that accentuates income gap between rural and urban areas is industrial productivity, which is often used to show how low the agricultural productivity is (see Table 2). An astonishing feature of Thailand is the contrast between the high labor share and low GDP share of agriculture. For example, in 1990, 64 percent of labor force produced only 12.5 percent of GDP. This means very low labor productivity in the agricultural sector, which is measured by per capita GDP, implying very high productivity in the non-agricultural sector. The gap reached as high as 12.4 times. Since the non-agricultural sector is concentrated in Bangkok and its vicinities, this productivity gap roughly corresponds to regional gap.

However, it must be noticed that this kind of statistic tends to exaggerate the productivity gap. One reason is that labor input is counted in terms of persons rather than hours worked. Farmers usually allocate only a part of their time to agriculture, especially during slack season, while many of the workers in the non-agricultural sector work full time in the sector. If productivity is measured in terms of hours worked, the gap will be smaller than this. Another reason is the tendency to overestimate non-agricultural sector, which will be discussed in the next section.

The decrease in the gap in the 1990s is impressive, which reached to 6.9 in 2000, though it is still high by international standard. The increasing income gap up to 1990 is simply explained by the fact that the agricultural share in GDP rapidly decreased while its share in labor force did not decrease as fast as the GDP share. And in the same way, the decreasing income gap is simply explained by the fact that the agricultural share in labor force rapidly decreased while its share in GDP did not decrease as fast as the labor force share.¹¹ The Gini coefficient in the bottom low of Table 2 is calculated with the productivity data.¹² This means that the Gini neglects the inequality within each sector and must be lower than the Gini for household income distribution. If this is true, the figure of 0.544 in 1970 implies still higher inequality of income distribution. However, this is not true. This is another evidence that the productivity data exaggerates the income gap. Corresponding to the decrease in the productivity gap, the Gini decreased very rapidly in the 1990s to 0.34. This table clearly shows that Thailand passed the turning point of income inequality in this sense in the 1990s.¹³ 

II. Appropriate Indicators of Regional Income Gap for Migration

The opportunity for one's income to increase as much as eight times is obviously alluiring for those in the Northeast. But we know many cases that people return to their village in the Northeast, giving up their jobs in Bangkok. A reason is that migrants' decision making is not solely based on economic reasons. There are several patters of migration depending on the life cycle.¹⁴ Another reason is that the income gap is not actually so big as the macro data show. If the income gap is so big, many companies in Bangkok will relocate their factories in Bangkok and its vicinities to the Northeast looking for cheap labor but this is not the case. In spite of the government policies to promote relocation as a policy to remedy regional income gap, only a small proportion of factories have been relocated. This implies that wage gap between Bangkok and the Northeast may not be so large. It seems that such a big wage gap exists when people move so frequently and the transportation cost is low.

The inappropriateness of using aggregated data such as per capita GRP for our purposes is now very clear. People are watching the income level that they may earn in Bangkok, with which they compare their current actual income in their villages. Per capita GRP is a completely inappropriate indicator for this purpose because it includes not only wages but also profits and other kinds of income that are not distributed to laborers. Bangkok, which has large sources of these incomes, will have higher per capita income. This is certainly an aspect of regional inequality. This kind of equality requires equal distribution of industrial structure. However, since industrial location is largely determined by geographical conditions such as accessibility to ports, unequal distribution of industrial location is not necessarily unjustifiable.¹⁵

Another problem is that it promotes the idea that a larger share of the agricultural sector is an indicator of backwardness, which is a reason to pursue a more equitable industrial structure.¹⁶ This image is straightforward if we only look at productivity figures. But this image underestimates the economic standard in rural areas. We often simply assume that most of the people living in rural areas only engage in agriculture and that most farmers earn income solely from agricultural activities. But this is not the truth. The difference between 40 percent of the agricultural share in labor force and 80 percent of the population living in rural areas implies that a large number of people in rural areas engage in non-agricultural activities and that even farmers earn a considerable proportion of income from non-agricultural activities. Rural areas are in no ways identical to agricultural sector. The arguments as to whether rural areas could have absorbed unemployed people during the crisis often confuse the distinction between rural areas and agricultural sector. Even if the agricultural sector may not have the absorptive capacity, rural areas can.

Our concern here is the income gap for migrants rather than that of industrial structure. A more appropriate indicator of regional income gap is household income and per capita household income that migrants earn currently in rural areas and will have in the destination. Table 3 shows that the income gap between Bangkok and the Northeast in terms of household income is “only” 3 times, which is considerably lower than the GRP data show but still much higher than that in other countries. In terms of per capita household income, the gap is slightly bigger, about 4 times.

As shown in Table 3, this table also shows that the income gap between Bangkok and the Northeast began to decrease. The largest gap was observed in 1990 when the income of the Northeast was only 30.6 percent and 24.4 percent in terms of household income and per capita household income, respectively. Since then the gap has been narrowing, though very gradually. The turning point can be observed more clearly by focusing on per capita income, which is more appropriate indicator than household income because it eliminates the impact of household size. In terms of per capita household income it is very clear that the other three regions also passed the turning point in the early 1990s.

Even though the income gap may be narrowed, the gap seems to be still large enough for rural people to rush to Bangkok but this figure does not guarantee that rural people can earn income 4 times as much as that in rural areas only if they move to Bangkok. This is not because they may not be able to find jobs in Bangkok as Harris-Todaro model assumes¹⁷ but because their jobs are limited to certain categories such as production workers, sales workers, service workers, and so on. The income gap that is relevant for rural people's decision making is therefore the one between the income level of these categories of jobs and farm income. Table 4 shows how narrow the rural-urban income gap is from farmers' point of view. Farmers in the Northeast earn on average 2,269 baht and 1,732 baht for those who mainly own land and for those who mainly rent land, respectively. If they work in Bangkok as general workers and earn their average income, namely 3,916 baht, their income will increase 1.7 and 2.3 times, respectively. If they can find a job as production or construction workers, the increase will be 2.2 and 3.4 times, respectively. And if they are lucky enough to find a job as a clerical, sales or service workers, the increase will be 3.2 and 4.2 times, respectively. These gaps seem to be large, but it should be noted again that the distribution within each category matters. If temporary migrant workers can enter only the lower end of these categories, their income will be lower than the average of the category and the income gap will be smaller.

Another interesting point of this table is that the relative income where Bangkok is set equal to 100 (see lower part of Table 4) varies in a very narrow range irrespectively of the category. For example, for the Northeast it varies from 45.3 percent for “professional, technical and managerial” to 58.6 percent for “entrepreneurs, trade and industry,” excluding two categories of farm operators.¹⁸ The same pattern can be observed for other regions. These findings mean that there are not so much regional income differentials if we look at them by socio-economic class. Even the income of “entrepreneurs, trade and industry” lies in this range, which is surprising because this category has less-organized labor market and faces high risks of earning. It can be said that what makes income gap in Thailand very large is not the regional income differentials by socio-economic class but the income differentials among various socio-economic classes and the composition of the socio-economic class. Since the latter is mainly determined by geographical conditions, which is inevitable to a certain extent, inequality that should be paid more attention is the income differentials among socio-economic classes. More concretely, our question should be why farm income is so low.¹⁹

Furthermore, if we take into consideration price differences between regions, or if we measure regional gap in “real” terms, the regional income gap will be even smaller because cost of living, especially housing, transportation and food, in Bangkok is much higher. Unfortunately we do not have accurate estimates of regional price level but if we had and estimated the real income level, the regional income gap will probably remain. The remaining income gap after adjusting the price level may be explained as an incentive to move to Bangkok. If there is no regional income gap in real term, nobody will move to Bangkok. The purpose for rural people to go to Bangkok is to save money to send back to home. Therefore for the rural people a large income gap is a good chance to save money. Therefore a large regional income gap may be desirable for them rather than a matter of complaint, unless it is a result of exploitation of the rural sector by the urban sector.²⁰

If our discussion is correct, the regional income gap may not be excessive in the eyes of farmers who are making decisions to migrate to Bangkok. This is, in this sense, in equilibrium, though we do not deny the fact that the farmers' income level in this equilibrium is still low in comparison with other socio-economic classes.

At the end of this section we examine the situation of poverty. Poverty incidence shown in Table 5 is based on a simple measure of the poverty line.²¹  This table shows that poverty incidence decreased very rapidly in every region in the 1990's. Even in the Northeast, it decreased from 48.4 percent to 19.4 percent in 1996.²² If this statistic is reliable, it can be said that the rapid economic growth in this period contributed to the decrease, which implies that the trickle-down effect was an important source of the decrease. In 1998, a year after the crisis, the poverty incidence increased slightly. If this statistic is correct, it may be said that the crisis increased poverty. However, we should notice that the poverty incidence in 1998 is much lower than that in 1994. It seems very doubtful whether these figures indicate that the poor were seriously affected by the crisis.

III. Situation of Labor Migration

In the previous section we have argued that appropriate indicators of regional income gap for migrants are not macro data such as per capita GRP but that the comparison should be between the income that the migrants actually earn in their villages and the income that they expect to earn in Bangkok. Since the migrants enter the lower end of labor market, the income gap is not excessive. On the other hand, it was shown that poverty in the Northeast is alleviated significantly. All these suggest that the image that extremely poor people move to big cities desperately may not be a common feature of migration in Thailand any more, though certainly many people are still forced to migrate because of extreme poverty.

Now we are going to examine how many people are actually moving and in which direction. Table 6 is calculated from the reports of the last four Population and Housing Censuses. In this table, migration is defined as “movement across villages or municipals in the period of 5 yeas before the census.” By comparing in-migration and out-migration, this table indicates that the Northeast has been always losing population by out-migration while Bangkok is growing by absorbing in-migration. The North and the South also have been losing population by out-migration to Bangkok. All the time migration has been occurred in a direction from the Northeast, North, and South to Bangkok and the Center. This is consistent with a popular image of migration. However, in reality the picture is more complex. People are moving in both directions as the numbers for in-migration and out-migration show. And people are moving to all regions, not simply to Bangkok and the Center. Many people in the Northeast move to the North and South to work as agricultural workers, fishermen, and so on. The reality is not so simple as the economic model.

A strange feature of Table 6 is that the number of migrants is too small. For example, the net migration of Bangkok is only 343,017 (=630,771-287,754) for a five-year period from 1985 to 1990, which means that every year on average only a net total of 68,603 people moved to Bangkok from the whole country. This is only one percent of the population of Bangkok. If it is compared with the natural growth rate of population, this size of migration may not be so great as to cause serious problems in Bangkok. Maybe this size of in-migration was needed for Bangkok economy to grow. It is better to understand that the figures in this table indicate long-term or permanent migration, rather than farmers' seasonal migration which is our focus.

More accurate data of migration is the Migration Survey conducted in 1997 (Table 7). An interesting point of this table is that the number of net migrants of Bangkok is -575,127, which means that Bangkok is losing this number of population by net out-migration. To the contrary, the Northeast received 364,671 people in net, most of which came from Bangkok. This direction of migration is exactly the opposite of what we have believed. An explanation of this phenomenon is that many of the seasonal migrants who went to Bangkok in dry season returned to the Northeast in rainy season, which is consistent with the fact that the survey period was in September.²³

We can see the seasonal change of employment in the agricultural sector from the Labor Force Survey as shown in Table 8. In 1998, employment in agricultural sector increased from 11.6 million in dry season (February) to 16.5 million in rainy season (August) by 4.8 million people while 2.1 million employments decreased in other sectors, of which 734,000 and 762,000 employments decreased in manufacturing and construction sectors, respectively. This implies that out of 4.8 million people who are engaged in agricultural sector and stop working in dry season, 2.1 million people are working in other sectors. The maximum possible value for migration to Bangkok is 5 million but it is unrealistic. Maybe only 2 million will move and the remaining 3 million stay in rural areas rather than being unemployed in Bangkok. If 2 million people will move between regions seasonally, this figure seems to be consistent with the 3.1 million of the Migration Survey that includes all migration within a province, within a region and between regions (see Table 9). A large part of the net out-migrants from Bangkok (575 thousand people in Table 7) can be considered to move to Bangkok in dry season and return to their villages in rainy season. 575 thousand people are equivalent to 10 percent of the population of Bangkok. If this figure is correct, Bangkok is fluctuating very regularly every year. On the other hand, 575 thousand people is only one percent of the country's population. The number of migrants who go to Bangkok from the Northeast (364,671) is only one or two percent of the population in the Northeast. We have to keep these figures in our mind when we discuss migration in Thailand and, especially the absorptive capacity of rural areas.

IV. The Economic Crisis and Labor Migration

It is widely believed that the economic crisis in 1997 hit the Thai economy so hard that everybody suffered from it. However, devaluation does not necessarily mean a negative shock to all people. For those who engage in export, it was a chance to make a profit because devaluation increased domestic prices of export products. Agricultural sector was also benefited from the increase in domestic price, though it is another question whether the benefit reached to farmers. Export industries were also benefited unless they depended on foreign loan excessively. However, voices of those who were in trouble were always louder than voices of those who were making a profit, which made people believe that everybody was suffering.

In these circumstances in addition to the traditional idea of migration, people imagined that poor migrants had to return to their troubled village. However, in the villages, life went on as though nothing had happened. Young people who returned from Bangkok spent idle time waiting for the recovery of the Thai economy. 24 They were not willing to take the so-called 3D (dirty, dangerous, and difficult) jobs even if the foreign workers from Burma, Cambodia and Lao P.D.R. who took these jobs were sent back to their home countries. Thus they became unemployed in rural areas, perhaps voluntarily.

Assuming that rural people were poor and suffered a lot from the crisis, international organizations, ODA institutions, NGOs and researchers came to the Northeast, the “poorest region” in Thailand. They had to find poor people who severely suffered from the crisis for their projects or for their thesis. It is usually not easy to contact those who are really poor because they do not want to show up and do not want to be recognized as poor. But there are many people who know very well that they can benefit a lot if they are recognized as poor. They are clever enough to answer the question as if they were really poor. Based on this kind of survey, many reports were published, most of which emphasized how people were suffered from the crisis. And perhaps many of them were written without visiting any villages, referring to some field research which were also based on a very few samples.

The Office of Agricultural Economics has been conducting a survey on farm household income periodically. The comparison between 1995/96 and 1998/99 indicates that farm cash income increased significantly except for the South where cash income decreased by 20 percent mainly due to the decreasing agricultural income.²⁵ Increase in farm income was largely due to the increase in agricultural prices caused by the devaluation of baht in 1997.

However, some crops were damaged by El Nino, so the positive impact of devaluation was offset to some extent. The decrease in farm income was attributed to the crisis, although El Nino had nothing to do with the crisis. Furthermore some papers even denied the capacity of rural areas to absorb unemployed people in urban areas.²⁶ This denial was needed in order to stay consistent with the idea that the agricultural sector is backward and poor. For them the agricultural sector should be poor to attract attention of people. This view was supported by the official statistics on poverty incidence, which increased from 11.4 percent in 1996 to 12.9 percent in 1998 in the whole country (Table 5). In the Northeast it increased from 19.4 percent to 23.2 percent in the same period, which may seem to evidence a “very serious” impact of the crisis. But the figure was much better than that in 1994, namely 28.6 percent. In other regions the situation was more or less the same. If this fact was taken into consideration, we may not have worried about poverty so much.²⁷

Once the target was set, a large amount of money was poured into the “poor” regions. Many kinds of funds were mobilized. The Japanese name Miyazawa became very famous in rural areas because a fund was called by the name of a Japanese prime minister at that time and Thai people confused other funds with Miyazawa. It was supposed that a lot of money was wasted for unnecessary projects such as road construction, which disappeared soon.²⁸ It was also said that the money just passed through the hands of villagers to reach to the ailing financial sector.

Following the guideline of the IMF, the Thai government had adopted an austerity budget policy until Thaksin took office. An austerity policy is always unpopular and the opposition party attacked the government on this point. When the opposition party took the office in 2001, it began to expand government budget bidding for popular support. Their constituency was coincidentally the “poorest” region and the “second poorest” region of Thailand. On the other hand, the South, the constituency of the opposition party and where Muslims dominate, is bypassed because their income level was widely believed higher than the North and the Northeast, though the poverty incidence in the South is actually higher than that in the North. As mentioned above, the image of poverty was created and exaggerated after the crisis, irrespective of the reality. The South might be more severely affected by falling commodity price in the world market.²⁹ 

Conclusion
At the end of 1997, the year of the crisis, a famous economist said to one of the authors that the Thai economy would never recover to five percent growth rate in the following decade. She seemed to be desperate because she could not do anything for her country to avoid the crisis. Many of the people in Bangkok were desperate and they thought everything very pessimistic, as if everybody was suffering from the crisis. To the contrary, it was a chance for foreigners to buy Thailand cheap, which was called “fire sales.” Many experts were sent to Thailand to help people, even though they knew nothing about Thailand. Economists who had little knowledge of financial sector were sent to Thailand as financial experts. They explained why the crisis occurred with the wisdom of hindsight, as if they had been long expecting the crisis to occur. Some economists came to Thailand to “analyze” how poor people were affected by the crisis. They collected statistics to show how Thai people suffered from the crisis. They looked only at one side of the matter and wrote many reports which exaggerated the seriousness of the crisis. Young researchers also came to Thailand with the prejudice that Thailand must be poor and became poorer after the crisis, read these reports and convinced themselves of the prejudice. Statistical analysis of poverty was conducted and obtained just another result. In order to stress the seriousness of poverty, a new poverty line was drawn. Field survey was conducted very briefly, and concluded that the Northeast was poor because of low productivity of agriculture, and proposed irrigation projects for example, even though the low productivity had nothing to do with the crisis. ODA organizations referred to these reports and proposed infrastructural projects and researchers “estimated” the benefit of the project to support it. Everyone takes advantages of this “opportunity” to protect their vested interests. They do not care who were really suffered from the crisis.

In this process, the traditional image of migration was revived and reinforced. This paper, however, does not insist that poverty is not a problem in Thailand any more but that we might have overlooked the poor who are really in need.

Endnotes

 1 The authors are grateful to Dr. Isra Santisart, Department of Economics, Chulalongkorn University and anonymous referees for their useful comments. Of course, the opinions expressed in this paper are the authors' own.

2 Dr.Yukio IKEMOTO is a professor of economics, Institute of Oriental Culture, the University of Tokyo, Japan.

3 Ms. Izumi TAKEI, is a research fellow of Japan Society for Promotion Science, Japan.

4 There are a few exceptions. See Isra Sarntisart, “Poverty Problem during the Economic Crisis” (in Thai), Thammasat Economic Journal Vol. 16 No. 3 September 1998; Behrman, J. R. and Tinakorn, P. “The Surprisingly Limited Impact of the Thai Crisis on Labor Including on Many Allegedly ‘More Vulnerable' workers,” Thai Development Research Institute(TDRI), 2000.

5 If they had a foreign debt, it means that the debt in terms of baht will be increased by devaluation.

6 Based on interviews to manufacturing companies in Bangkok. At first it was believed that income inequality would decrease because it was expected that the urban rich were most severely hit by the crisis. But this was not the case. Clever people are always clever and could make a good profit even during the crisis.

7 When people need aid, they tend to pretend that they are seriously suffered. But this attitude allows outsiders to intervene them excessively. In the case of Thailand, foreign agents such as international organization, foreign governments, foreign companies, and NGOs, made use of this situation.

8 As the neglect of the South region in terms of poverty and migration, see Institute for Population and Social Research (IPSR), Mahidol University, Migration and the rural family: sources of support and strain in a mobile society, 1997. This neglect may be one of the reasons for the recent turmoil in the South.

9 In economic theory, equilibrium is achieved when income gap disappears and migration stops. In reality, even when equilibrium is achieved, people will migrate in both directions and income gap will remain due to price gap and other differences.

10 In the dual economy model, workers migrate to urban areas so long as urban wage is higher than rural one. Higher wage in urban areas is caused by scarcity of labor force in urban areas, while lower wage in rural areas reflect abundant labor supply in rural areas. As the migration continues, labor force will become scarce in rural areas while it will be abundant in urban areas, which means income gap will narrow and the incentive to migrate will disappear.

11 This may seem to be strange if you are used to the idea of decreasing marginal productivity, which means that when labor input becomes larger, the marginal product will be smaller. However, here we discuss the relative share of agricultural sector, not the absolute amount of products.

12 The Gini coefficient ranges from 0, perfect equality, to 1, perfect inequality. The Gini coefficient of household income distribution for relatively equal countries is about 0.3 while that for relatively unequal counties is bout 0.5.

13 The turning point of income inequality, so called the Kuznets' hypothesis, was discussed in Ikemoto Yukio and Uehara Mine, “Income Inequality and Kuznets' Hypothesis in Thailand,” Asian Economic Journal Vol. 14 No. 4, Dec. 2000 pp.421-443.

14 Migration to and from Bangkok may be classified into three patterns as follows based on our field survey:

(1) Young man and women move to Bangkok to work just after graduating from high school and after a several years return to hometown for marriage.

(2) Married men and women go to Bangkok to work during agricultural slack season.

(3) Married men and women stay in Bangkok for a longer period while their parents take care of their children.

15 This is not to deny the effect of industrial relocation on rural income but it is not a necessary condition. There are many other ways to achieve the same goal.

16 The national account statistics tend to overestimate “manufacturing” sector by classifying agricultural processing to the manufacturing sector, even though the process may be a very primitive one.

17 See Harris, J. R. and Todaro, M. “Migration, Unemployment and development: A Two-Sector Analysis” The American Economic Review, Vol.60, 1970. In the case of Thailand most of migrants secure job in the destination through human network before they migrate so that they will not be unemployed as the Harris-Todaro model assumes.

18 These two categories are excluded because agricultural sector in Bangkok is very exceptional.

19 In the age of globalization, cheap imported staple foods hit domestic agriculture hard, pulling down agricultural prices and pushing out agricultural workers to urban sectors, both of which contributed industrialization. Low farm income, cheap wage labor and high profit rate in industrial sector, all of which mean high income inequality, are interconnected at the bottom of the economy.

20 Many people from “poor” developing countries come to Japan to save money to send back to their countries. For them, the bigger income gap between Japan and their country is more desirable. For them, Japan is not a place to live permanently but to work and save money. After several years of stay in Japan, they are very happy to go back to their country to enjoy a life with their family.

21 Deficiencies of poverty line measure are not limited to theoretical ones which have been discussed in academic papers. A more serious problem is that it can be very easily manipulated by adjusting poverty line. It is usually adjusted by price level but we do not have accurate data of regional price level. The relevant price level for poverty line is the one which reflects the consumption basket of the poor but it is not separately estimated. Therefore it contains a considerable margin of errors, which means the poverty incidence can be manipulated within this margin. Anyway we must be very careful in interpreting the poverty incidence.

22 As for the increase in 1998, which is just after the crisis, we need more careful examination before we say something about the impact of the crisis. A reason is mentioned in the previous note. We will discuss this later.

23 The impact of the crisis might not be large, taking into consideration the survey period, from September 1995 to September 1997.

24 It was mainly younger generation that lost job after the crisis. See, World Bank, Thailand Social Monitor; Thai Workers and the Crisis, pp.26-28, 2000.

25 This data from the homepage of Department of Agricultural Economics, Ministry of Agriculture and Cooperatives (http://www.oae.go.th/statistics/index.html).

26 Employment in agricultural sector increased in 1998 and 1999 by 0.6 million and 1.1 million, respectively. On the other hand, the net outflow of employment from agricultural sector in the dry season decreased by 56% in the period of three years from 1996 to 1999. The net outflow to service sector dropped by 104%, which means that employment in service sector moved to agricultural sector even in dry season. These figures seem to clearly prove the absorptive capacity of agricultural sector. See World Bank, Ibid.

27 There appeared many reports which emphasized, sometimes exaggerated, the seriousness of the impact. Japan Bank for International Cooperation, Poverty Profile: Thailand (in Japanese) p.2, 2000.

28 A news paper reported that a road disappeared after a few days of heavy rain.

29 The price of rubber decreased very rapidly in the latter half of the 1990s. There might be a sense of inequality or injustice behind the recent turmoil in the South.