The welfare analysis in the previous lectures 13 and 14 measured the income effects of government intervention in food markets in terms of producer and consumer surplus and changes in government revenue. Transfer efficiency analysis is concerned with the magnitude of the transfer losses associated with agricultural support instruments. This lecture has three objectives:
Support instruments have two main effects: they create transfers which reflect changes in the distribution
of income and they impose economic costs on the economy.
Transfers are payments from one agent in the economy to another for which there is no corresponding flow of goods
and services. The transfers created by agricultural support instruments can be calculated without the use of model
assumptions, based solely on accounting data and price information. Since transfers are always from one agent to
another agent, the sum of transfers between producers, consumers and taxpayers due to agricultural policy instruments
is, by definition, zero. The OECD's Producer Support Estimate and Consumer Support
Estimate are examples of indicators which calculate transfers in this way.
Transfers due to support policies generally differ from the income effects measured by the producer and
consumer surplus and government revenue changes in welfare analysis. The transfers would be equal to the income
effects only if the support instruments creating the transfers did not effect the behaviour of the agents in the
economy (thus creating distortionary costs) and if they were not associated with transaction costs (see below).
Consider a small open economy which imposes a tariff on imports in order to support farm producers. Production and demand after the tariff are Q4 and Q3 respectively. On the basis of these observed quantities and the two observed prices, we define the following:
The economic costs for the economy of support instruments may be divided into distortionary costs and transaction
costs.
Distortionary costs are created when support instruments create a (price) wedge between the marginal rates of transformation
of different agents in the economy. These costs can be calculated as the sum of the real income effects for farmers,
consumers and taxpayers created by the government intervention and are represented by the deadweight costs identified
as B and D in the above diagram. Note that these real income effects and hence distortionary costs cannot be directly
observed. They can be calculated only with the help of models representing economic behaviour. Such models are
not always available, and, when available, are often based on controversial assumptions.
Transaction costs are costs other than the distortionary costs associated with government intervention. These costs
include the costs of policy implementation, the costs of producing, collecting and processing information, and
the costs of control and enforcement. In the EU, intervention costs are an important example of transaction costs
which are not captured by, and which are additional to, the deadweight distortionary costs.
Deadweight costs and policy implementation costs are examples of direct costs associated with the policy. In addition,
there can be indirect costs. For example, indirect distortionary costs arise where taxes have to be imposed
in order to finance the policy. Where the marginal social cost of taxation is greater than the value of the revenue
raised, then the budget cost of the policy should be raised to reflect this higher cost. Such indirect costs might
be positive as, for example, when tariff revenue is collected which may be used to reduce even more distortionary
taxes elsewhere in the economy.
Finally, indirect transaction costs can be augmented by rent-seeking behaviour which implies that agents
other than the government use real resources to try to influence and capture the benefits of support. Such costs
include the lobbying costs of farm organisations, for example.
The following is a full tableau of the identified economic costs of a policy. Not all of these costs are necessarily
taken into account in empirical studies.
|
Direct costs |
Indirect costs |
|
| Distortionary costs |
Deadweight costs |
Cost of public funds |
| Transaction costs |
Administrative costs |
Rent-seeking costs |
The objective of transfer efficiency analysis in evaluating agricultural policies is to relate the combined taxpayer and consumer costs to the additional income which farmers receive. In terms of welfare analysis, transfer efficiency can be defined as the net income gain to farmers arising from a one unit gross transfer cost to consumers and taxpayers.
where 0 < TE < 1
Transfer efficiency is closely related to the cost of making a unit transfer, UTC. Consider the following expression for the deadweight economic cost of a transfer policy to farmers
If we now divide and multiply the RHS by 
,
and substituting in the expression for TE above, we obtain
By dividing across by 
, we obtain
This expression shows that the unit cost of making a transfer and the transfer efficiency are directly related.
If TE= 1, then there are no deadweight costs, while if TE = 0, the deadweight costs absorb all of
the transfer.
Note that this definition of transfer efficiency only takes account of the direct distortionary costs identified
above. A more complete analysis would include all of the costs identified in the tableau above as part of the denominator
of this expression.
There can also be a reduction in transfer efficiency if there are significant distributive leakages arising from the transfer. Distributive leakages arise if part of the transfer benefits groups other than the target group of the transfer. Such leakages might include benefits to upstream or downstream industries, benefits to non-target farmers (see Lecture 16 for information on the distribution of CAP direct payments), or benefits to third countries (where the changes in supply and demand induced by the policy affect the level of world prices, thus changing a country's terms of trade, see Lecture 17). While, apart from the last category, such leakages are not overall costs to the economy (in that someone has gained from the gross transfer from taxpayers and consumers, just not the target farm group), they do reduce the transfer efficiency of a given policy.
An important source of distributive leakage may be the capitalisation of farm programme benefits into the prices of inelastic factors of production. For example, suppose that land is fixed in supply. Higher support prices increase the profitability of production and will encourage farmers to seek to rent or buy additional land. The demand for land will increase until the profitability of farming is restored to its pre-support level. If all farmland was rented, this suggests that farmers would be indifferent to the level of support, as all changes in farm profitability would be reflected in the price paid for this inelastic input. Another example of a fixed factor of production is quota rights. Expanding dairy farmers must purchase or lease quota from exiting farmers, and much of the benefit of milk price support transfers to the quota owners, the exiting farmers, rather than supporting the incomes of working dairy farmers.
If empirical analysis suggests that the transfer efficiency of agricultural policy is low, then considerable
gains might be achieved by improving transfer efficiency. The point of departure for transfer efficiency analysis
is a desired level of farm income. Consequently, neither the motivations underlying the decision to transfer income
to farmers nor the determinants of the size of the transfer are assessed.
Students should be aware of a number of issues when undertaking cost and benefit studies of transfer policies:
European Commission, 1994, 'EC agricultural policy for the 21st century', European Economy, Reports and Studies No. 4, Brussels, Directorate-General for Economic and Financial Affairs. See Chapter B ' The economic costs of agricultural policy'. (Lecky 338.1094N42).
OECD, The incidence
and income transfer efficiency of farm support measures, AGR/CA/APM(2001)24/FINAL
(also available as Part II of OECD, 2003, Farm Household Income: Issues and Policy Responses, Paris, OECD (pp. 53-84) (NOTE: 2MB file to download)
(it is not necessary to work through the mathematical formulae for calculating transfer efficiency for different types of
agricultural support instruments; read the introduction and conclusion and examine the four diagrams for the different types of support instruments)
Supplementary reading
Brooks, J., 1996, Factors conditioning the transfer efficiency of agricultural support, OECD/GD(96)186, Paris, OECD.
(note that this is a pdf file but that the suffix is incorrect on the website. Download the English version of the Acrobat file to your computer and change the suffix from .eng to .pdf in order to be able to open it)
OECD (1995), Assessing the relative transfer efficiency of agricultural support policies and 'Transfer efficiency of agricultural price support; in Adjustment in OECD Agriculture: Issues and Policy Responses, Paris, OECD (not available in Trinity Library).
For a very accessible empirical application, see
Kola, J., 1993, 'Efficiency of supply control programmes in income distribution', European Review of Agricultural
Economics, 20, pp. 183-198.