In this article we will discuss about the model of labour migration and reallocation.

Employment policy in developing countries like India cannot be formulated and implemented without answering a basic question, viz., how can underutilised labour be used in a develop­ment strategy? Ragnar Nurkse and W. A. Lewis asserted that large numbers of people remain engaged in work which adds nothing to national output. Nurkse saw the reallocation of a sur­plus labour to more productive uses, especially labour-intensive construction projects, as a major source of capital formation and economic growth.

Lewis envisaged a similar reallocation process but he pictured the ‘capitalist sector’, essentially industry, as the principal employer of surplus labour. Both theories regarded the labour reallocation process as nearly costless but they worried about how to capture from the agricultural sector the food necessary to feed the transferred workers.

While criticising the Lewis model J. R. Harris and M. P. Todaro have developed a new model of economic development which is relevant for labour surplus countries like India. It is the best known model of internal migration in the context of present-day developing countries. The model has focused on migration of labour from rural to urban areas induced by certain incentives. They have referred to two types of migration—induced migration and internal mi­gration.


According to this model migrating workers are essentially participants in a lottery of rela­tively high-paid jobs in the towns. When new urban jobs are created the lottery becomes more attractive to potential migrants. Depending on their responsiveness to this improved opportu­nity, more than one worker are likely to migrate for each job created.

If so, the output foregone may be that of two or more agricultural workers, not just one. If the migrants bring some of their family members to urban areas more output will be lost. The reason is that the wife and children of migrants find fewer employment opportunities in towns than in the rural areas because they do not have land in the towns on which to grow food.

Internal Migration:

Most of the internal labour migration which occurs in the course of economic development is from rural to urban areas. The model sets out to explain why the observed high rates of small- urban migration found in most developing countries is quite natural from an economic view point.


In this model the expected real wage-differential between urban and rural areas is the main motivating force behind migration or the main determinant of the migration decision. The potential migrant first calculates the real income which he would get in the urban area for a job with his present effort. He next makes a personal judgement of the probability of obtain­ing such a job.

He then compares the expected income with that which he hopes so obtain in the rural area. His migration decision is based on the difference. For example, suppose the ex­pected rural income is Rs 1000 per month, and that real income of an urban job which is commensurate with his qualification is Rs 2000.

However, this information alone is not suffi­cient to make the migration decision. If the subjective probability of getting the urban job was 0.4, than the expected income would be Rs 800 and no migration would take place on this case expected urban wage = actual urban wage x the probability of getting a job

800 = 0.4 x Rs 2000


If the probability were 0.8 the expected income would be Rs 1600 (= 0.8 x Rs 2000) and the worker would migrate. (To this one may, of course, add the cost of transfer.) Thus the model brings into focus the role of economic incentive in the migration decision.

The Basic Model:

The Harris-Todaro model assumes that migration from rural to urban areas depends primarily on the difference in wages between the rural and urban labour markets.

That is:

where Mt is the number of rural to urban migrants in time t, is response function, Wu is the urban wage and W is the rural wage. Since there is unemployment in the town (and it is as­sumed that there is no unemployment in rural areas), and every migrant cannot expect to find a job there, the model postulates that the expected urban wage with the rural wage. The expected urban wage is the actual urban wage times the probability of getting a job, or

where Weu = expected urban wage and p = probability of getting a job Here p is expressed as

where Eu is urban employment, Uu is urban unemployment and L is total urban work force. Harris and Todaro assume that all members of the urban labour force have equal chances of obtaining the jobs available. So Weu becomes simply the urban wage times the urban em­ployment rate.


Migration in any given time then depends on three factors:

(a) The urban-rural wage gap,

(b) The urban employment rate and


(c) The responsiveness of potential migrants to the resulting opportunities.


where Mt = migration in period t and h = the response rate of potential migrants


As long as Weu > Wr the rural-urban migration will continue. It will stop only when migra­tion has forced down the urban wage or forced up urban unemployment so much that expected urban wage (Weu) is covated to existing rural wage (Wr). If Wr > Weu there will be a flow of disappointed urban jobseekers back to the rural areas. This is known as reverse migration.

Theoretical Implication of the Model:

There is no denying the fact that any development project can be evaluated using social cost- benefit analysis. An important part of the social cost of any input is opportunity cost, its value in its next best alternative use. Labour hired for an urban formal-sector project might well be drawn from the urban informal sector also.

The worker who moves out of the urban informal sector may, in turn, be replaced by someone from the rural sector. In this case, the output lost is that of the worker who was formerly in the rural sector, i.e., the worker at the end of the em­ployment chain. For this reason, some analysts believe that the wage paid to casual agricultural labourers provides a good measure of the social cost of unskilled labour.

However, this measure, although a good indicator of output foregone through labour reallo­cation, probably understates the true social cost of employing labour, which has other compo­nents that are likely to be significant. One such component is induced migration. No doubt there are both pull and push factors behind internal labour migration that occurs in the course of economic development from rural to urban areas.

Such migration can result either from favourable economic developments in the towns or from adverse developments in the rural areas. The Harris-Todaro model integrates these two sets of forces in their analysis of the proc­ess of labour reallocation that is likely to occur during economic development. This is why the model was sort of innovation in the literature of development economics when it appeared for the first time in 1970′.


The Policy Implications of the Model:

The H-T model has far-reaching implications from the policy point of view. For example, if the government of the country concerned were successful in fostering industrial development in an urban area, employment would increase there.

The effect would be to increase the subjective productivity of getting urban employment in the minds of rural inhabitants. Migration would increase and the eventual effect of the new industrial development could be that urban unem­ployment becomes higher that the level prevailing before the new development took place.

There will be some level of urban employment which ensures equilibrium in the sense that no further migration takes place. Potential migrants may take a long-term view in arriving at a decision. They may consider that their desireness of obtaining an urban job will be higher after a waiting period of some months. Thus they will compare the present value of the sum of expected urban earnings with that of expected rural earnings.

They may be content to accept a low wage in the urban informal sector for some time. This might be a rational decision on a long-term basis. The root of the problem is the large difference between earnings in the modern industrial sector and those in the rural areas.

Often the former are well above the market clear­ing levels for varies reasory. The long-term solution to the problem lies in adopting policies for both urban and rural areas which reduces the real income differences between the two areas.