A question of causality: Does modernization of agriculture lead to economic growth or does growth induce a modernization of the agricultural sector? For many years, this question has been hotly debated among development economists. While those economists who believe in growth-led agriculture (GLA) were dominating until recently, now the proponents of agriculture-led growth (ALG) are afloat again. Which insights does this debate yield for Georgia?


For a long time, the question seemed to be settled. If one asked a development economist in the 1960’s, by all likelihood s/he would have opted for GLA. This consensus derived from the observation that the Industrial Revolution in England, as well as the subsequent economic developments in major Western countries, seemed to be GLA stories. The Industrial Revolution clearly started in the wake of certain innovations in the industrial sector, most notably the steam engine, which were not caused by modernization in agriculture. To the contrary, little development could be observed in agriculture throughout the whole Middle Ages. Except minor changes, a farmer in the year 700 was using the same technology as a farmer in 1700. When industrialization picked up, its blessings spilled over to the agricultural sector, not vice versa.

While many economists read history in this way, it became a building block in the theoretical edifice of the father of development economics, Albert O. Hirschman. Hirschman argued that developing countries should embark for a strategy of unbalance growth. Instead of developing across the board, countries should fully concentrate on strategic sectors of their economies which would then tug the other sectors out of backwardness. For deciding which sector was strategic, Hirschman’s concept of “linkages” came into play: the criterion was the potential of a sector to influence other sectors of the economy. An example of a sector which would qualify as strategic under this criterion is the steel industry. If a country developed a steel industry, this would lead to a demand for raw materials, for labor of various qualification levels, and, through the demand for labor, also for food. These were called the backward linkages, because they stimulated earlier stages in the value chain. Most importantly, raw materials, food, and labor can be provided in a developing country, so that the demand of the steel industry would be met by domestic supply. At the same time, a steel sector would provide the basis for producing more sophisticated goods like cars and all kinds of machines, and these effects further down the value chain were called forward linkages. According to Hirschman, it would be inefficient to develop the agricultural, steel, and machine sectors simultaneously in a protracted and lengthy process. Rather, one would entirely focus on the steel sector and let the linkages do the rest.

Under this paradigm, agriculture has no great significance due to its lack of substantial linkages. In an underdeveloped country, improving the conditions for agriculture would not create demand for anything the country can already deliver, e.g., if modernization would create a demand for tractors, these would have to be imported. And even the forward linkages of agriculture are ambivalent. If more food is available, there will be more people, yet a lack of people was never the bottleneck for growth in a developing economy. To the contrary, higher birth rates usually aggravated social and economic problems.

This view was reflected in concrete policies. As Awokuse and Xie (2014, Canadian Journal of Agricultural Economics) write: “Policy makers in many developing countries proposed and adopted development strategies that were antiagriculture. […] In many developing countries, the agricultural sector was subject to heavy taxation. For example, prior to agricultural reforms in 1979, Chinese agriculture was under a heavy tax burden and the revenues were used to subsidize urban and industrial development.”


Since its inception, the “linkages” paradigm of Albert O. Hirschman lost much of its appeal. This may be due to the overall failure of development economics, which was dominated by Hirschman’s ideas for a long time (this overall failure has not been resolved so far – Rajan and Subramanian (2008), Review of Economics and Statistics, report the stunning (and devastating) result that there is no empirical correlation between aid received by a country and its growth rate). Leaving Hirschman’s ideas behind and forgetting about the Industrial Revolution, there are indeed reasons why agricultural development might precede overall growth. Most importantly, if agricultural produce is exported, it may provide the capital that is necessary to fund industrial growth – an important aspect in developing countries, which are, like Georgia, notoriously short on capital. Likewise, the increased supply of domestically produced food may lead to import substitution if a country, like Georgia, covers a considerable share of its consumption through imports. This allows for the use of scarce capital for more productive purposes than importing food. Finally, if agricultural modernization is achieved through the accumulation of human capital, the overall quality of labor may improve to the benefit of other sectors. For example, if farmers learn to manage their businesses, keep contracts, make plans etc., these skills will also be useful for non-agricultural employment.

Free of any ideology, Awosuke and Xie (2014) estimate econometrically whether the ALG or the GLA model better describes reality, and they find that in some countries one can observe ALG and in others GLA. In their sample of nine countries, development follows a GLA-pattern in Brazil, Chile, Mexico, Kenya, and South Africa, and it follows an ALG-pattern in China, Indonesia, Thailand, and Cameroon.


UNDP and ISET have recently concluded a study that projects structural developments in Georgia’s agricultural sector in the upcoming 5-15 years. According to our study, without policy interventions, more than 46% of Georgian farmers will quit agriculture in that time span. Yet, this is not the Industrial Revolution story. Georgian farmers do not leave agriculture because there are plenty of jobs in manufacturing pulling them to the cities, but mostly because they are old (average age in the sample was 55 years) or their agricultural operations do not yield substantial incomes. Georgia seems to be bound to an ALG development path due to the mere lack of other sectors which could tug agriculture out of its backwardness.

If Georgia develops according to the ALG-pattern, the Georgian agricultural sector will develop without preceding growth of the other sectors of the economy. However, ALG can be influenced by smart policies. For example, agricultural development can be focused on labor-intensive branches of agriculture (e.g., berries, walnuts, wine), so that the increased productivity, which will inevitable make some farmers leave agriculture, will not drive so many people out of agriculture. Labor-intensive products are often relatively high-valued, and their exports have the potential to generate capital inflows that can be used for industry development. Smart policies would also aim at developing industries which take agricultural produce as input, extending the existing value chains within the country. This would quite literally create agriculture-led growth. Moreover, modernization of agriculture should go along with upgrading of the skills of the rural population, so that also in this respect, agriculture provides the basis for growth in other sectors.

Georgia is not in an economically favorable position due to its geographical location, which is all but advantageous, its lack of technological leadership in any field, its relatively expensive labor (compared to many Asian countries), and the lack of neighbors that heavily invest in the Georgian economy. In this situation, agriculture may be the only development alternative that remains, and Georgia should steer this process actively. As sketched in the diagram, smart agricultural development in Georgia should revolve around economic viability, labor intensity, environmental friendliness, innovation, and knowledge. Clearly, these aspects are differently pronounced in animal and plant production. Yet, what do these terms imply concretely? We will discuss the way forward for Georgian agriculture in future articles that will appear in this place.

See our article usage guidelines