To stop taxing agriculture, governments have to do more than dismantle the interventions in agricultural prices---they have, in addition, to eliminate other taxes on agriculture, reducing the protection of industry and getting the exchange rate in line with its long-run equilibrium value
Industry has been the darling of development, certainly for the now-industrial countries. Following suit, those with the reins of policy in developing countries decided that agriculture was impervious to price incentives, so they believed that taxing it would sacrifice little in output---or so went the conventional wisdom.
Suspecting that the conventional wisdom was wrong, Maurice Schiff and Alberto Valdes---in a large comparative study that they and Anne O. Krueger directed---examined how price interventions affect agricultural growth and overall economic growth.[1] They measured the income transfers that price interventions induced among agriculture, government, and the rest of the economy. A seemingly straightforward line of inquiry, but no small task for a 25-year period in 18 countries.
Another piece of the conventional wisdom was that taxing agriculture was easy to administer and extremely attractive in countries with a thin tax base. In addition, shifting scarce resources to industry was thought to be justified by the rising protection in industrial countries and by agriculture's declining terms of trade---a pound of agricultural exports was buying less and less than a pound of industrial imports. So, policymakers taxed the daylights out of agriculture, secure that they were doing the right thing. (Taxation here is used broadly and includes the burdens of industrial protection and overvalued exchange rates.)
By the 1980s the conventional wisdom was beginning to look less wise. It was becoming apparent that far more than agricultural price policy was influencing the decisions of farmers to invest and to produce ---the likelihood of unsavory shifts in the exchange rate, for example, was a powerful impediment to investment. It was also becoming apparent that taxing agriculture was sacrificing far more in output than the earlier surmise.
Governments influence agricultural prices both directly, through agricultural sector policies, and indirectly, through industrial protection and macroeconomic policies that tax agriculture relative to tradables and nontradables outside the agricultural sector.
What have been the effects of such direct and indirect interventions in this study's 18 developing countries over 25 or so years? On average, they taxed agricultural producers by about 30 percent.
Schiff and Valdes found that the indirect effects (22 percent tax) were by far the stronger, easily outweighing even the positive influence of direct interventions in the few cases where direct measures protected agriculture. Macroeconomic policies caused the appreciation of the real exchange rate and raised the relative cost of nontradable inputs and reduced the real purchasing power of income received from the sales of export and import-competing commodities. And protection for domestic industry hurt agriculture by raising the domestic price of importable agricultural inputs above world prices, by reducing the purchasing power of farm households as consumers of manufactured goods, and by causing further appreciation of the real exchange rate.
Many direct measures have also taxed agriculture---on average, by 8 percent---depressing the prices received by agricultural producers. Governments intervened directly through procurement measures (government marketing boards are often the only legal buyers of agricultural outputs), quotas on exports of food crops and other agricultural commodities, and direct taxation of such exports.
In most industrial countries, the main objective of agricultural price policies is to maintain farm income and employment in the face of declining real world prices for cereals---with massive net income transfers to agriculture. In most developing countries, however, the primary objectives have been food self-sufficiency, domestic price stability, low food prices for urban consumers, and government revenue. On average, the net effect of direct and indirect interventions has been an enormous income transfer out of agriculture--- averaging 46 percent of agricultural GDP a year over 1960---84.
An analysis of these income transfers---and of who gained and who lost from them---helps to reveal some of the motives for price interventions and to identify the forces arrayed against reform. Agriculture was clearly the loser, while the big winners were government (net revenue gain), urban consumers (lower food prices), and industry (cheap raw materials and other inputs).
To examine the impact of price policy on annual growth of real agricultural GDP, Schiff and Valdes compared the average agricultural growth rate in the group of countries in which nominal taxation rates were lower (nominal protection rates were higher) than the average with the rate in the group in which taxation rates were higher (protection rates were lower) than the average. The group with the lower taxation rates (higher protection) showed a higher average growth rate.
The authors also used a regression estimate of the growth rate of real agricultural GDP to simulate its average growth rate in the absence of total price interventions. Here, too, the relationship between total taxation (protection) and agricultural growth was significant: the lower the tax on agriculture, the higher the growth. The higher prices in agriculture reduce labor outmigration from the sector, increase investment, and encourage wider adoption of new techniques---and result in a higher growth rate.
The high tax on agriculture in many developing countries has been motivated in part by the belief that industry was the dynamic sector while agriculture was static and not very responsive to incentives. So, economic growth would be accelerated by shifting resources from agriculture to industry.
Several of the country studies explicitly mention such an industrialization strategy: Argentina, Chile, Colombia, the Dominican Republic, Ghana, Malaysia, Morocco, Pakistan, and Zambia. And most of the others suggest that such a policy was implicit in increasing government revenue by taxing agriculture and using it to increase spending outside agriculture.
These policies did not accelerate growth, however. Schiff and Valdes examined the relation between price interventions and overall economic growth in three ways. First, they compared the various measures of intervention with the rates of growth of GDP across four groups of countries. Second, they used regression analysis to relate the various measures of intervention to GDP growth for the 18 countries. Third, they examined the relation between the various measures of intervention and GDP growth for the 18 countries, dividing them into two groups based on the average size of the various measures of intervention.
All three tests suggest a negative relation between the rate of total taxation of agriculture and GDP growth: policies that depress agriculture's terms of trade below international levels are associated with slower economic growth. The authors also found that GDP growth rose as the indirect taxation of agriculture fell across the four country groups.
The findings provide a solid base for prescribing agricultural price policy in the developing countries. Schiff and Valdes write that if you want to prosper, don't tax agriculture relative to other sectors by protecting industry and maintaining overvalued exchange rates. But if you stop taxing agriculture, you have (in many countries) to look to other sources of revenue to finance the activities of government.
It is also essential within agriculture to stop taxing exports and protecting imports (to put imports and exports on an equal footing)---and to dismantle quotas, licenses, state trading companies (which obscure the real winners and losers from subsidies or taxes), and internal agricultural marketing regulations that prevent a free flow of goods and services within the country.