The optimal level of investment in agricultural innovation (Part 2)

The optimal level of investment in agricultural innovation

-

Johannes Roseboom, Innovation Policy Consultancy, The Netherlands

Email: j.roseboom[at]planet.nl

-

Policymakers not only have to make decisions about whether government support is needed, but also about how much support should be given and in what form (tax deduction facilities, subsidies, or direct provision of services).

The most common way of evaluating the level of government investment in agricultural innovation is that of benchmarking. However, despite its widespread use, benchmarking is a rather poor tool because it lacks any theoretical underpinning. It tends to reinforce the status quo, which may not be optimal at all. For example, many economists have argued (based on accumulated evidence of hundreds of ex post rate-of-return studies of agricultural research and extension projects) that there is serious underinvestment in agricultural innovation.

For the theoretical answer, the following simple economic model can be used to establish the optimal level of investment in agricultural innovation. It takes three steps, namely:

1. Calculate, using a standard cost-benefit analysis technique, for all innovation project proposals, the expected rate of return (ERR). For a portfolio of agricultural innovation project proposals, the distribution of these rates will most likely take the form of a declining slope – there are only a few project proposals with a high ERR and many with a low or negative ERR.
2. Select among the projects that target the same problem or opportunity the one with the highest ERR (i.e., eliminate any duplication of effort).
3. Of the remaining projects, finance all projects with an ERR higher than the minimum rate (as set by the public or private financer). The optimal budget can now be calculated as the sum of the budgets of all projects that have passed the minimum rate.

In practice, however, the selection of agricultural innovation projects hardly ever takes place in this strictly rational, economic way. In the public sector (both in developing as well as developed countries) it is certainly not common practice to calculate for each agricultural innovation project proposal an ERR.[1] Instead, project proposals are at best being evaluated and ranked on the basis of multiple criteria – many of which, however, reinforce some form of economic rationality. The resulting ranking of project proposals is not perfect, but one could still accept it as a reasonable approximation of the economic ranking. The biggest problem, however, is that such ranking does not provide for a clear cut-off point. It is the available budget that determines how many projects will be funded and one does not know whether this budget is on target or either under- or overshooting the optimal investment level.

Another distortion that is quite common is to deviate from the ranking in order to achieve a politically acceptable spread of selected projects across commodities, topics, geographical areas and executing agencies. In this process, higher-ranking proposals are traded in for lower-ranking ones, reducing the overall profitability of the portfolio of selected projects.

Difference between the public and private optimum

Depending on whether a public or a private perspective steers the evaluation of a portfolio of project proposals, the outcome (in terms of the projects selected and optimal budget allocation) will differ. There are two factors that cause this difference:

1. The minimum cut-off rate used by the private sector (about 15-20%) tends to be substantially higher than the rate used by the government (roughly 6% in developed countries and 12% in developing countries). This can be related back to the fact that borrowing money is usually more expensive for the private sector than for the public sector. In addition, the private sector adds in a substantial profit margin on top of the interest rate.
2. In the calculation of the ERR of an innovation project, the benefit stream as identified by the public sector includes both producer and consumer benefits, while the benefit stream as identified by (an agency acting on the collective behalf of) the private sector only captures the producer benefits. The costs of the projects are in both scenarios the same. As a result, the same innovation project captures a higher ERR when evaluated from a macro-economic point of view (the central planner’s perspective) rather than from a micro-economic point of view (the private enterprise perspective). Moreover, the relative ranking of projects will be altered as well – projects generating predominantly consumer benefits will drop in the private ranking. All-in-all, one can expect the private optimum investment level to be lower than the public one. This insight is an important argument in support of subsidizing private innovation in a market economy. From a private perspective, such subsidy lowers the costs of innovation projects and hence increases their ERR. As a result, more project proposals will pass the private cut-off rate, reducing the difference between the private and public optimum investment level.

The supply of innovation projects

The economic selection of agricultural innovation project proposals as presented above only applies to proposals that have been submitted for evaluation. In that sense, the calculated optimum is the optimum given the human (in terms of researchers, extension specialists, etc.) and physical innovation capacity in place. Decisions about building such capacity cannot be steered by individual projects, but are of a long-term, strategic nature. For example, in order for a country to gain from modern biotechnology in agriculture it has to make a substantial upfront investment in human resources and physical infrastructure before any concrete biotechnology project proposal will see the light of day. Such investments require considerable strategic foresight in understanding in which direction science will develop and how it could be applied for the benefit of the country. The amount of resources that a country can afford to invest in its (agricultural) innovation capacity depends to a large extent on the level of economic development of a country. Financial assistance from donors can help to moderate this constraint to some extent.

At the same time, however, there are also various structural factors (often correlated with the stage of economic development) that influence the overall profitability of agricultural innovation and hence the size of the optimal investment. Understanding these structural factors may help to identify and pursue policies that could improve them. Some of the more important structural factors include:

1. The level of technological knowledge within the agricultural innovation system. Policies that can enhance this factor include investment in basic science, education (both academic and practical), and access to knowledge (e.g. investment in ICT infrastructure).
2. The level of risk and uncertainty. In order to reduce risk and uncertainty, policies should be geared towards: (a) overall political and macro-economic stability; (b) clarity and consistency regarding intellectual property rights, ethical standards, product standards, environmental standards and other regulatory measures of relevance; and (c) the development of capacity to predict future developments (technology foresight studies, scenarios, roadmaps, etc.).
3. Weak economies of scale. Policies that could have a positive impact on this factor include: (a) strengthening of collective action within agriculture and agricultural value chains by providing legislative and financial support; and (b) supranational collaboration.
4. Weak efficiency and effectiveness. Adopt policies to improve the management and organization of the lead actors (agricultural research, extension, farmer organizations, etc.) in the agricultural innovation system.
5. Monopolistic behaviour. Private or state monopolies lack the economic incentive (i.e. competition) to innovate. Policies that lead to the opening up of agricultural input and output markets should bring the necessary competition into the system for innovation to take off.
6. Weak rural institutions and infrastructure. Lack of credit, lack of markets, and high transportation costs are often mentioned as important constraints when it comes to innovation in agriculture in developing countries. Policies that target these constraints help to boost the profitability of agricultural innovation activities. It is important to move in all these areas simultaneously in order to create the highest level of synergy.

What all these structural factors have in common is that they tend to be correlated with the stage of economic development. The more advanced the economy, the more favourable these structural factors become. In other words, the optimal agricultural innovation budget is not the same for all countries (e.g., expressed as a percentage of agricultural GDP), but tends to increase with economic development. This suggests that just investing a lot in agricultural innovation capacity at an early stage of economic development (and then just hoping that a lot of profitable innovation opportunities will pop up) is not going to resolve the problem. It can only work when the structural factors are being improved as well.

Conclusion

In this section we have outlined how the optimal level of investment in agricultural innovation can be calculated using a fairly simple model. This optimal level only applies to innovation activities for which the economic impact is the principal objective (which is usually the large majority of the investments). Basic research falls clearly outside this definition because its principal objective is to advance fundamental knowledge.

In order to operate the economic model, it is necessary to make a cost-benefit analysis (and derive an ERR calculation from it) for each project proposal. The reality is that this is not standard practice in the public sector (yet) and hence this will continue our relative ignorance about the optimal level of investment in agricultural innovation.

The selection methods used instead (such as multi-criteria) are less than perfect, but should still steer us towards selecting the more promising innovation projects.

[1] There tends to be a perception that calculating the ERR of an agricultural innovation project is rather difficult and costly. To a large extent, this is the result of a lack of expertise and experience in making such calculations. The advantage of making a calculation is that it forces people to quantify relevant variables. This very much helps to reveal underlying assumptions and uncertainties. In the private sector, the use of cost-benefit analysis and selecting (innovation) projects on the basis of their ERR is a far more common practice.