Which one of the following topics would be of most interest to a public choice economics?

1.1 Basic rules of the game of public choice and of Public Choice

Public choices are not something made “in public”, at least not necessarily so. Nor are these choices made “by the public” in any literal sense. Literally speaking “the public” is not a choice making entity. What renders “public choices” public are their potentially wide ranging external effects which tend to concern the public at large. But that does not amount to the presumption that there is a specific choice making entity whose involvement would separate public from other kinds of choices. Quite to the contrary like on markets, in the realm of politics, results emerge from choices of persons (or, perhaps, agents) but are not choices made by any person.

On markets and in politics individuals interact with each other according to certain “rules of the game”. In this wide sense the concept of a “rule of the game” is taken as comprising causal laws (natural) as well as man-made (artificial) rules. As opposed to the Hayekian usage of terms which treats spontaneously emergent rules as “natural” (see [Hayek, 1973/1993]) rules need not be deliberately enacted to qualify as “artificial”. The crucial point is that they are entities that would be otherwise if men acted otherwise (see related to this [Heinimann, 1987/1945; Buchanan, 1979]). Game theoretically speaking the so-called “rules of the game” comprise everything that is beyond the causal influence of the decision making entities in the game (thus including preferences and artificial as well as natural features of the situation as long as they are beyond the strategic influence of the players in plays of the game considered).

Adopting the broad use of terms as suggested by game theory we can say that if we could take together all the rules of the game (all that is beyond the strategic influence of actors in the game) we would get a characterisation of the “game of life” in full. According to this view “public choice” is a particular “part” of “the game of life”. It is the game we are playing when we “play politics”.

Since the behaviour of human individuals is always what it is, human behaviour, Public Choice insists on using the same model of behaviour throughout. Whether it be political or other behaviour, its explanation should be based on a universal explanatory model for which the “rules of the game” are antecedent clauses while the model of individual behaviour as such must be the same across games (which of course does not rule out that some rules of a lower order game are to be explained as emergent or artificially created in a higher order game).

Beyond the requirement of universality of the basic explanatory model of individual behaviour which seems to me impervious to criticism Public Choice traditionally makes the additional assumption that neo-classical homo oeconomicus is in fact the empirically correct universal behavioural model. To the early theorists of Public Choice who mainly came out of economics this assumption may have seemed rather natural. But it is a much more precarious assumption than that of the universality of the behavioural model.

The Theory of Public Choice

The theory of public choice developed by James Buchanan and his followers challenges virtually every tenet of mainstream public sector theory. Buchanan described in his Nobel lecture delivered in Stockholm, Sweden, in 1986 the foundations of the public choice perspective.4 The disagreements with the mainstream view begin at the most basic level with assumptions about how people behave. According to Buchanan, the mainstream theory assumes that people are essentially schizophrenic. They are self-interested in their economic lives, but they suddenly become other-interested and consider the broader social or public interest in efficiency and equity when they turn to the government in their political lives. Nonsense, say public choice advocates. People do not change their stripes; they remain self-interested in their political lives as well. They turn to the government only because they cannot get what they want for themselves in the marketplace, and they view the government as just another venue for seeking their own objectives. Buchanan refers to individuals’ interactions with the government as fiscal exchanges to mirror the self-interested motivations of standard market exchanges. Using the government in the pursuit of self-interest is viewed as entirely appropriate and legitimate.

The thrust of public choice theory is positive, not normative. Buchanan scoffs at the notion of an idealized, beneficent government acting as an agent of the people in pursuit of social objectives. Instead, Buchanan argues that public sector economists should be studying actual political and governmental institutions and determining whether they give people what they want. The test of government efficiency in this positive vein is simply how well the government serves each person’s self-interest. Full efficiency requires unanimity under democratic decision-making because only then will no one lose as a result of any government policy. This is as “efficient” as the government can be in helping people get what they want. Notice that the public choice definition of efficiency in political activity is far stronger than the economic definition of efficiency as Pareto optimality, which the mainstream perspective uses to judge public policies.

The public choice perspective has normative content but is strictly process-oriented and concerned only with the rules that govern political activity. Moreover, Buchanan claims that normative content centers on a single point in time—at the founding of a democratic nation. The norms are embedded in the constitution drafted by the nation’s constitutional convention.

When focusing on the constitution, Buchanan was influenced by the Swedish economist Knut Wicksell, who theorized about the legitimate role of a democratic society at the end of the 19th century. Wicksell first thought of government activity in terms of fiscal exchanges and described the ideal as unanimous consent for all policies at every point in time. Buchanan concedes that requiring unanimity all the time is asking for too much because it would lead to paralysis. Instead, he points to the constitution and argues that legitimacy in government requires only a consensus among the framers of the nation’s constitution about the rules under which the government is permitted to operate. When designing these rules, the convention members think only of their self-interests and those of their descendants as they perceive them. Unanimous agreement at the constitutional convention about the rules of politics is the ideal, although Buchanan concedes that a consensus may be all that is possible.

The only valid normative test of government activity at any time after the convention is the following: Could the current rules that guide and constrain government activity have arisen from an agreement at the constitutional convention? If the answer is yes, then the current rules are legitimate, and society has forged a legitimate link between the people and their government. Notice that the policies that result from these rules cannot be evaluated directly by any norms. In particular, policy outcomes are irrelevant in and of themselves. According to this test, process is everything, namely, consistency with the self-interested rules agreed to at the constitutional convention.

Normative policy analysis after the convention is possible but is limited to suggestions for constitutional reform and then only if the normative test fails. Normative proposals take the form of recommending changes in the constitutional rules such that people are better able to pursue their self-interests in their fiscal exchanges with the government. For example, Buchanan seriously doubts that the large, prolonged U.S. federal budget deficits that have existed in most years since the early 1980s would pass his normative constitutional test because of the damage they could inflict on future generations. He favors a balanced-budget amendment to the constitution.

An interesting question is whether redistributional policies or rules could ever achieve a consensus at a constitutional convention, given that redistributions force some people to pay taxes for the benefit of others. Those who are taxed may well feel that they are not getting what they want from their fiscal exchanges. Buchanan believes that consensus could be reached if the framers of the constitution choose to consider the welfare of future generations and are willing to view the future through a veil of ignorance. The idea is that no one can predict the future; therefore, no one at a constitutional convention can know with certainty how their descendants will fare for all time. Therefore, they may see establishing rules that permit redistributions of income on the chance that their descendants might be the ones who fall on hard times as being in their self-interests. In other words, they are simply allowing for the possibility of future transfers to their families.

The public choice perspective is persuasive in a number of respects. The assumption of self-interested political behavior is instinctively appealing to economists, and much political behavior is clearly self-interested. The insistence on analyzing actual political institutions and actual political choices is also sensible, as is a focus on the constitutional rules that guide and constrain all political activity. Nonetheless, public choice has not captured the day among public sector economists but remains a distinctly minority perspective if the weight of the professional literature is an accurate guide.

Perhaps the mainstream has stood firm against the public choice challenge because the normative basis of public choice theory is so thin. The public choice perspective, as articulated by Buchanan, lacks any clear sense of good citizenship or empathy—qualities that many people believe are essential ingredients for a society that anyone would want to live in. A narrow focus on self-interested constitutional rules may not be enough to sustain a comprehensive normative economic theory of the public sector. In any event, the majority of economists apparently want to directly judge the results of specific government policies and to do so in terms of the Pareto efficiency criterion and commonly accepted equity norms, such as equal opportunity or horizontal equity. More generally, government activity motivated entirely by self-interest simply does not have the normative appeal of government activity motivated by the public interest in efficiency and equity.

The battle between public choice and mainstream economists is unlikely to be decided on empirical grounds because ample evidence exists to support both sides. Two published reflections by Joseph Stiglitz and Joel Slemrod are instructive.5

Stiglitz, a Nobel laureate, has contributed as much as any economist to mainstream public sector theory during the past 50 years. When he was asked to reflect on his years at the Council of Economic Advisors, he responded with a paper describing why the government has such difficulty enacting policies that are so clearly beneficial from the mainstream perspective. The problem in a nutshell, according to Stiglitz, is that all too many government officials behave as Buchanan said they would. They pursue and protect their self-interests rather than the public interest, such as by keeping their private information secret when it is in their personal advantage to do so. Stiglitz believes that the government is hugely beneficial overall but not nearly as much as it could be if officials were more consistently public-spirited.

Joel Slemrod has been a major contributor to mainstream tax theory and policy during the past 40 years. He speculates that other-directed, civic-minded behavior may produce much more than just a kinder and gentler society. He points to some studies that show a positive relationship between economic growth and prosperity and what he terms social capital—such things as the degree of trust in others, the propensity to obey society’s rules, and civic behavior. The social capital variables in these studies are obtained through surveys. A connection among civic-minded, other-directed behavior and economic growth will be a major boost for the mainstream perspective if it stands up to further analysis.

Conclusion

The authors have considered three economic perspectives on public choice that help to explain why it has often proven difficult to obtain political backing for apparently common-sense public health interventions. The health and societal implications of many public health interventions can never be assessed in their entirety in advance of implementation. Even if they could, however, public health advocates have often found themselves in conflict with powerful interests groups, and politicians or bureaucrats who pursue their own objectives. This introduces an important and complex set of constraints into the priority-setting process, implying that available funds might be spent in particular areas or on specific programs determined by informational advantages or power structures within society. In general, such constraints will result in departures from conventional criteria such as cost-effectiveness rules.

Is it possible to counter those powerful influences? History has plenty of success stories. From the second half of the nineteenth century, Britain's major cities embarked on a reform of the municipal social health amenities and social services that resulted in significant improvements in health. Historians believe that a class-bridging coalition between grass root organizations of the growing urban population, a new generation of civic leaders with social conscience that originated from the well-off urban elite, and a strong cadre of public service professionals, notably Medical Officers of Health, secured these reforms. Similar developments happened in other European countries and the USA.

There are modern day examples of spirited initiatives by governments, government departments, or community organizations that instigated radical improvements in public health. For example, the outbreak of the plague in the Indian city of Surat in 1994 led to a decisive reorganization and introduction of stringent performance management of the civic waste department, named ‘transformation from AC to DC’ – making bureaucrats leave their ‘air conditioned’ offices to the ‘daily chore’ of direct supervision of waste management on site. The participatory budgeting model introduced in 1989 in the Brazilian city of Porto Alegre was an innovative reform program that successfully overcame severe inequality in living standards among city residents. Part of the program was introduction of participatory budgeting; community members now decide how to allocate part of the public budget. The recent decisive ruling of the Australian government on plain packaging laws for cigarettes is celebrated as a great success in the fight against cigarette smoking. For decades, the tobacco industry has successfully prevented such laws, to protect the value of their brands' image, which they used more or less openly to link the brand with popular causes or films – such as the placement of Phillip Morris brands in Superman I and II, see Figure 6.

The authors have merely scratched the surface of this underresearched area of study. There is great scope for a much better understanding of decision-making behavior – both in low- and high-income countries – and a range of hypotheses can be tested by examining the priority-setting process itself and the resulting patterns of healthcare and public health expenditure. They would not challenge the desirability of seeking to maximize the health gains of the health system through the use of CEA. However, health economists have not yet taken advantage of the full range of economic approaches at their disposal. Only by securing a better understanding of the decision-making process can the impact of CEA be enhanced. They would argue that this can be achieved by augmenting the understanding of the political context of priority setting, using a variety of well-established models of political economy.

Public Insurance

Redistributive transfers motivated by a desire for income insurance are consistent with the public choice perspective. Buchanan argued in his Nobel address that the framers of a nation's constitution might permit redistributive public insurance programs such as social security pensions and unemployment insurance if they choose to view the future behind a veil of ignorance in which the future is truly uncertain (Buchanan, 1987). This vantage point raises the possibility that some of the framers or their descendants may become impoverished, in which case allowing for redistributive public insurance can be viewed as self-serving.8 The demand for public or social insurance is considered in Chapters 20 (medical care) and 21 (public pensions) of this text.

Public assistance can also have an insurance motive, at least partially. Thomas Husted attempted to distinguish between altruistic and insurance motives for public assistance in the United States on the basis of survey data collected as part of the 1982 American National Election Study (Husted, 1990). The participants were asked whether spending on food stamps and on AFDC was too much, about right, or too little. Husted hypothesized that the motives for food stamps were likely to be purely altruistic, with accountability. In contrast, the motives for AFDC were likely to be a mixture of altruism and insurance because the majority of the spells on AFDC are very short term, often only a month or two. Using econometric techniques suitable for survey responses, Husted obtained estimates that support his hypotheses. The demand for food stamps was uniformly upward sloping in income, whereas the demand for AFDC was U-shaped in income. An upward-sloping relationship between public assistance and income is consistent with an altruism motive. The inverse relationship between AFDC and income at the lower incomes is consistent with an insurance motive among the near-poor.

Husted's interpretation of the insurance motive is reasonable but open to question. One wonders how the poor and near-poor were able to muster support for insurance-based transfers as they tend to have little political influence. A possibility is that the rich also support public assistance at least partially for its insurance properties as Buchanan had suggested, but that the regression equation cannot separate insurance and altruistic motives among the rich.

Public Insurance

Redistributive transfers motivated by a desire for income insurance are consistent with the public choice perspective. Buchanan argued in his Nobel address that the framers of a nation’s constitution might permit redistributive public insurance programs such as social security pensions and unemployment insurance if they choose to view the future behind a veil of ignorance that leaves the future as truly uncertain (Buchanan, 1987). This vantage point raises the possibility that some of the framers or their descendants may become impoverished, in which case allowing for redistributive public insurance can be viewed as self-serving.9 The demand for public or social insurance is considered in Chapters 20 (medical care) and 21 (public pensions) of this text.

Public assistance can also have an insurance motive, at least partially. Thomas Husted attempted to distinguish between altruistic and insurance motives for public assistance in the United States on the basis of survey data collected as part of the 1982 American National Election Study (Husted, 1990). The participants were asked whether spending on food stamps and on AFDC was too much, about right, or too little. Husted hypothesized that the motives for food stamps were likely to be purely altruistic, with accountability. By contrast, the motives for AFDC were likely to be a mixture of altruism and insurance because the majority of the spells on AFDC are very short term—often only a month or two. Using econometric techniques suitable for survey responses, Husted obtained estimates that support his hypotheses. The demand for food stamps was uniformly upward sloping in income, whereas the demand for AFDC was U-shaped in income. An upward-sloping relationship between public assistance and income is consistent with an altruism motive. The inverse relationship between AFDC and income at lower incomes is consistent with an insurance motive among the near-poor.

Husted’s interpretation of the insurance motive is reasonable but open to question. One wonders how the poor and near-poor were able to muster support for insurance-based transfers because they tend to have little political influence. One possibility is that the rich also support public assistance at least partially for its insurance properties, as Buchanan suggested, but that the regression equation cannot separate insurance and altruistic motives among the rich.

5 Conclusions

This chapter has focused on the economic foundations of technology policy, a subject that necessarily involves consideration of additional fields including management issues in both the public and private sectors and issues of public choice outside economics. At the outset, a “benchmark” theory based upon Arrow and Nelson's arguments regarding the potential underperformance of markets in supplying adequate amounts of investment in new knowledge was constructed. In modifying this framework to address issues of technological as opposed to scientific knowledge, the centrality of intellectual property and appropriability issues was identified. While variants of this benchmark are commonly employed in governing technology policy initiatives, it was argued that the political salience of technology policy has led to policy outrunning theory. Developments in economic theory related to sectoral imbalances in productivity advance, the implications of asymmetric holdings of information, and the role of localization in the generation and distribution of knowledge have provided opportunities to reconnect policy with theory.

Reflections on these theoretical developments and practical experience provided a basis for developing a “taxonomy” of technology policy designs. This taxonomy was employed to organize a number of specific analytical conclusions regarding the relation between the “practice” of technology policy and its theoretical foundations. In Section 4, these designs were considered from the perspectives of planning, implementation, and evaluation. A principal observation of this section is that assessment of sponsor and performer capabilities is of critical importance for effective policy planning. In implementation, the potential for reproducing the features of patent races through coincident or strongly overlapping designs was noted, a possibility that suggests the need for further consideration of technology policy “strategy.” Crafting technology policy strategy, however, remains a problematic task due in large measure to the pervasiveness of technological issues throughout government and the resulting plethora of interests vying for priority in any such policy strategy. Developing a more systematic account of what policy initiatives have been undertaken and expanding the literature on international comparisons of such initiatives would be an advance over current tendencies toward ad hoc policy planning.

The large area of program evaluation methods and experience was not considered in this chapter. A systematic analysis and history of developments in program and project evaluation would allow it to be better integrated with the range of economic theories considered in this chapter. The dichotomy between “benchmark” and alternative policies chosen to organize this chapter has somewhat obscured the role of evolutionary and institutional economics which have made important contributions to both the issues of knowledge localization and information asymmetry. Many of the developments related to innovation systems are derived directly from evolutionary economics critiques of conventional economics assumptions of widespread and persistent market equilibrium. Evolutionary economists have also directly attacked the conventional assumption that knowledge is a global public good, noting the difficulties of translating between information and knowledge. Evolutionary economics has also stimulated renewed attention to issues of variety generation and selection processes which are closely related to issues of path dependence, increasing returns, and transitions between different social and market “paradigms.”

In the light of past contributions that have attempted to link economic theory and technology policy, the main aim of this chapter has been systematically to trace the sources of economists’ skepticism regarding the theory and practice of technology policy and to demonstrate how these have evolved over time. The “benchmark” theory is still routinely employed in public discourse and in the governance of new initiatives. Newer approaches not only widen the scope for intervention, they also suggest the need for greater planning and evaluation capabilities to increase the chances of success for such initiatives. In normative terms, technology policy should be built upon firmer theoretical and empirical foundations rather than relying upon ad hoc expertise or political predictions. Constructing such foundations is the principal challenge for this area in the coming years.

2.5 Economic Regulation

‘The Theory of Economic Regulation’ (1971b) attempts to answer that question. It is one of a series of contributions by Stigler to what has since come to be called ‘public choice’ economics: the shift from viewing the political market as not susceptible to economic analysis, as one in which disinterested politicians and bureaucrats pursue the ‘public interest,’ to viewing it as one in which the participants are seeking, as in the economic market, to pursue their own interests, and hence subject to analysis with the usual tools of economics. The seminal work that deserves much of the credit for launching ‘public choice,’ The Calculus of Consent, by James Buchanan and Gordon Tullock (1962), appeared in the same year as the Stigler– Friedland article.

The ‘central thesis of the article,’ Stigler wrote, ‘is that, as a rule, regulation is acquired by the industry and is designed and operated primarily for its benefit.’ He notes that two ‘alternative views of the regulation of industry are widely held. The first is that regulation is instituted primarily for the protection and benefit of the public at large or some large subdivision of the public. The second view is essentially that the political process defies rational explanation.’ He then gives example after example to support his own thesis, which is, by now, the orthodox view in the profession, concluding: ‘The idealistic view of public regulation is deeply imbedded in professional economic thought. The fundamental vice of such a [view] is that it misdirects attention’—to preaching to the regulators rather than changing their incentives (quotes from Stigler 1971b, p. 3).

‘Smith's Travels on the Ship of State’ (1971a), published in the same year as ‘The Theory of Economic Regulation,’ raises the same question on a broader scale. Smith gives self-interest pride of place in analyzing the economic market, but does not give it the same role in analyzing the political market. Smith's failure to do so constitutes Stigler's main—indeed, nearly only—criticism of the Wealth of Nations, that ‘stupendous palace erected upon the granite of self interest’ (Rosenberg 1993, p. 835). The same theme pervades many of Stigler's later publications.

8 Political economy angles on underinvestment in agricultural R&D

Over time and among countries, we observe a remarkably consistent pattern of very high measured rates of return to investments in public R&D, which we interpret as evidence of systematic, sustained and substantial government failure with high social opportunity costs. Why is it so? It is not so easy to answer this question well and confidently, for reasons we explain next, but some studies have sought to do so using public choice models.

Drawing on foundations laid by luminaires such as Becker (1983), Krueger (1974), Peltzman (1976), and Stigler (1971), over the past four decades agricultural economists have applied various public choice models that link agricultural policy choices to their consequences for influential (or otherwise favored) interest groups.45 In these models, public agricultural policy is driven by its distributional implications, which can potentially account for the observation that, rather than correcting for market failures and increasing aggregate welfare without regard for the distributional implications (as in the Kaldor-Hicks approach to normative applied welfare economics, which underpins the idea of “government failure”), in many cases agricultural policy conversely reduces aggregate economic welfare in the process of redistributing it. This alternative conception is demonstrably relevant for accounting for the various farm commodity price and income support programs imposed by governments around the world. Interest group models are also likely to be relevant for understanding both public investments in agricultural R&D and the regulation of agricultural technologies.

The “political economy” of agricultural and food policies has been the subject of many papers and several reviews, including the survey by Anderson, Rausser, and Swinnen (2013) and the book by Swinnen (2018a), which together cover most of the previous work. As the review by Anderson et al. (2013) reveals, the work in this area offers considerable conceptual insight and much useful measurement of the disaggregated consequences of policies, but much less empirical depth and detail on the specific causes of policies.

The authors show how agricultural support policies vary systematically among and within places. Primary determinants of the size of transfers to agricultural producers are the stage of economic development (with transfers to farmers and other interest groups tending to increase with increases in per capita incomes and reductions in the share of agriculture in the economy) and the trade status (transfers tend to be greater for importable goods than exportables, especially for large-country importers). These broad outcomes are easily rationalized. Less insight (and even less empirical evidence) is available as to why certain producer groups are subsidized more than others, or why certain instruments of policy are favored in particular settings. The reasoning tends to become somewhat circular: to account for the observed patterns it must be the case that the more heavily subsidized groups have stronger political influence, and that the chosen policy instrument is more efficient in some sense.

Positivistic studies of agricultural and food policies have been mostly focused on farm commodity policies and trade policies, but agricultural R&D policy has received some attention, both directly and in terms of its interaction with and implications for other policies. These studies use information and ideas about the likely distributional consequences to draw inferences about policy choices. In applications to agricultural R&D, typically untested (and largely untestable) modeling assumptions become crucial for determining findings concerning the distributional outcomes that drive the consensus political economy narrative.46 In particular, many studies arbitrarily impose a pivotal research-induced supply shift in a closed economy model (with inelastic demand for farm products) such that they find consumers necessarily benefit and producers necessarily lose, by assumption!47 Hence, the authors infer, farmers are opposed to agricultural R&D. But we have no evidence that it is reasonable to assume a pivotal shift as the default. And it would be much more empirically reasonable to employ an open-economy model in which demand for farm products from any country is elastic—in most cases, highly elastic—such that farmers would benefit along with consumers even from a pivotal supply shift (e.g., Alston, 2002a, 2002b, 2018).

Nor do we have any substantial evidence that farmers as a group generally take the view that productivity enhancing agricultural R&D is inimical to their interests and that they consequently are opposed to new technology and public and private investments in agricultural R&D. We can readily provide evidence of the converse case in which farmer organizations lobby for public research spending and organize to tax their own products to finance R&D programs (e.g., see Alston et al., 2003, 2004; Alston & Fulton, 2012). Opposition to agricultural technologies seems more likely to come from groups representing consumers, especially in rich countries, even if it is readily apparent that the technologies do make food more abundant and cheaper for consumers. These observations seem to fly in the face of the logic of the typical political economy models that predict consumers will support and producers will oppose agricultural R&D. It seems more reasonable to presume, as a baseline, that both producers and consumers in any country are unlikely to lose and that both are more likely to benefit from agricultural R&D—as in Alston, Norton, & Pardey, 1995.

All these studies typically say little if anything about the distribution of consumer benefits among consumers, or of producer benefits among producers, which might also matter. While all consumers stand potentially to gain from research that makes farm commodities more abundant and cheaper, the distribution of the benefits among them will depend on various attributes that differ among consumers, depending on where they live and what they consume; cheaper meat does not benefit vegans! Among producers, only those who adopt a resulting innovation stand to benefit from the research, and non-adopters will be made worse off by any resulting fall in output price or increase in the price of inputs they use (see, e.g., Alston, 2002a, 2002b and Alston, 2018). Among agribusiness firms, the issue turns on whether they provide goods and services that are complements to, or substitutes for, farm commodities made more abundant and cheaper by agricultural research.

More complex treatments allow for the interaction between (and possible joint determination of) agricultural R&D investments and other farm policies. For example, de Gorter and Zilberman (1990) propose a model of industry technology with inelastic demand wherein farmers would lose from research in an undistorted setting but would benefit from research in the presence of a target price policy. Thus, they suggest we can account for and justify farm support policies as having been introduced to make possible socially beneficial research that otherwise would not have been politically acceptable to agricultural interests. Using the same kinds of modeling assumptions, Swinnen (2018c) derives the same kinds of results concerning the structural connections between other (exogenous) attributes of the market and policy context, and jointly endogenous research policy and other farm policy. Again, as with the discussion of research in isolation, the work is predicated on untested and unlikely modeling assumptions that imply dubious distributional details, which are central to the analysis; it is entirely theoretical.

Direct empirical evidence on the determinants of investments in agricultural R&D is scanty. In a comparatively early contribution, Alston and Pardey (1993, 1994) investigate whether agricultural price distorting policies had contributed to low rates of productivity growth in agriculture by causing an underinvestment in agricultural research. This pair of closely connected papers provides (1) a broad overview, critical assessment, and synthesis of the relevant previous work into a coherent framework, (2) a synopsis of theoretical and empirical findings concerning the ways in which price distorting policies influence the size and distribution of research benefits, (3) drawing on that synopsis, a theoretical (interest group) model of the determinants of investments in agricultural R&D, including some detailed comparative statics and a nuanced discussion of the implications, and (4) a regression analysis using extensive international data, modeling the determinants of agricultural research intensities.48

As the authors explain, theory is ambiguous about the links between national investments in public agricultural research and rates of assistance to agriculture, for various reasons beyond issues about the nature of the research-induced supply shift, detailed above: “While the weight of the analysis favors a positive relationship between the rate of protection and the size of the research budget, the opposite result cannot be ruled out” (Alston & Pardey, 1994, p. 76). Their regression models fit the data reasonably well, and all of the coefficients on the economic variables are statistically significant and plausible (see, e.g., Alston & Pardey, 1994, Table 4.7 and related text, pp. 78–80). Specifically, they find that agricultural research intensity (i.e., spending on agricultural research per dollar of agricultural output, ARI) increases with (1) increases in total agricultural output, (2) increases in income (GDP/capita), (3) increases in the share of agricultural production exported, (4) decreases in agricultural exports as a share of total exports, (5) increases in agricultural spending per unit of output, and (6) decreases in the rate of protection to agriculture.

The key result is the last in the list: everything else equal, higher agricultural research intensities are associated with lower rates of protection. It is robust to variations in details of the specification and even to the use of alternative data on the protection rates. It holds both when a single total measure of protection is used or when protection is partitioned into direct and indirect elements. This finding contradicts many of the theoretical models: price support policies would seem to serve more as a substitute for, than as a politically enabling complement to, agricultural R&D spending. These results might not have any particular policy implications, though they do cast into question some studies that purport to do so.

Agricultural R&D policy entails institutions governing not just the total amount of R&D funding but also (1) the sources of the funds (e.g., public versus private, and within public, various agencies of federal, state and local government) for particular research projects and programs, (2) the emphasis of research priorities across fields of science, among commodities, across parts of the supply chain, among geographic locations, and between agriculture and the environment, (3) the financial and other institutional arrangements governing the division of labor and distribution of funds among different research performers—whether done by block grants, competitive grants, or on some other basis, and (4) any attendant restrictions on how the funds may be used and how the research may be undertaken. Alston and Pardey (1996) and Alston, Pardey, and Roseboom (1998) discuss these “financing” aspects of U.S. and global agricultural R&D policy from a normative perspective. Clearly there is scope for bureaucratic and other rent-seeking, regulatory capture, and other forms of government failure in all these dimensions, and opportunity for economists to employ political economy models to analyze the phenomena.

A case in point is the CGIAR which has seen a rise in earmarking by donors, supporting projects that will benefit their own constituencies at home, as a concomitant of increased funding support; they have virtually eliminated the unrestricted core funding that was the hallmark of the early halcyon days of the system (Alston et al., 2006, 2020). These political constraints on the use of funds almost surely reduce the rate of return to the portfolio, but are integral to the phenomenal past increases in the size of the portfolio—trading off the intensive margin against the extensive margin. On the other hand, the resulting inefficiencies might have contributed to the recent funding collapse. Similar stories might be relevant for the shifting sands of public funding for national agricultural research programs in countries like the United States (see, e.g., Alston & Pardey, 1996; Pardey et al., 2010, 2013a; Pardey & Smith, 2017).

2.3 Social Determinants of Motives

Social modification of motives and actions is usually dealt with in contemporary social science by introducing evolutionary mechanisms (Nelson and Winter 1982). In most current versions of Darwinism, the sole effective motive is to increase fitness (essentially rate of population growth); altruism will not compete. This has often been interpreted as implying that only selfish human motives survive, and that, consequently, maximization of utility means maximization of income or wealth, profit of the business firm, or power (the latter in Public Choice Theory in political science and sociology). In this view, only the special genetics of ants and bees, and some other social insects, allow nonpropagating offspring to exhibit altruism to close kin.

However, it has been shown (Simon 1997) that, in the presence of bounded rationality, altruism may increase the fitness of altruists, and, hence, may survive and grow within a society. If bounded rationality admits a large capability for social learning, persons willing and able to learn (docile), will have a large fitness advantage over others. By natural selection the society will help individuals acquire knowledge and beliefs that usually enhance their individual fitness, but are sometimes advantageous to the society's fitness at the individual's expense (e.g., volunteering for military service). As long as social influence, on balance, increases individual fitness, altruism can also thrive, contributing to the fitness and growth of the society (see also Altruism and Self-interest).