1. Introduction
James Buchanan was a longstanding advocate of federalism. Contrary to some accounts, he did not first defend federalism in the 1950s, in response to the desegregation debates occurring in Virginia; rather, he defended this specific form of democracy in the 1940s, at the very beginning of his career. As Marianne Johnson (Reference Johnson2014, Reference Johnson2018a, Reference Johnson and Wagner2018b) has shown, federalism was at the core of his Reference Buchanan1948 dissertation, Fiscal Equity in a Federal State.
But even this was not Buchanan's first work on federalism. The first and earliest form of Buchanan's views on federalism can be found in two term papers he wrote while he was a graduate student at the University of Chicago – “‘Federalism’: One Barrier to Labor Mobility” (Reference Buchanan1946b) and ‘A Theory of Financial Balance in a Federal State’ (Reference Buchanan1947).Footnote 1 These essays are shorter, thus less detailed and, in some respects, less rich than the dissertation they predate. There is nothing on the voluntary exchange theory, the Italian public finance economists, Lindahl, or Wicksell. There is nothing on the neo-Jeffersonians, the neo-Hamiltonians, the possible tyranny of the majority on a minority, or even the “rules of the social game.” Yet most of the ideas Buchanan would develop in his dissertation on federalism – and in academic articles published later –are there.Footnote 2 This is why these essays are important. The primary objective of this paper is to discuss and analyze them and to enrich our understanding of Buchanan's views on federalism.
Using these two essays, we show that Buchanan had already identified the inequality that exists among states in a federation, an inequality that arises because different regions have different economic resources and therefore different fiscal capacities. He had already noted that centralization would remove but not solve the problem. The solution would consist in using a principle of justice – an equal treatment of equals – equalizing unconditional grants and interarea transfers. In addition, Buchanan had made some claims that are not in his dissertation, but to which he returned later (see e.g. Buchanan, Reference Buchanan1950, Reference Buchanan1952, Reference Buchanan and Buchanan1961), even using them against Charles Tiebout and inter-jurisdictions competition (see Boettke and Marciano, Reference Boettke and Marciano2016; Buchanan, Reference Buchanan1957). One set of claims involved labor mobility. Buchanan argued against mobility, because it would generate more inequality and injustice in an institutional framework in which inequality and justice already existed. In a federation with rich and poor regions, migration would impoverish poor regions and the impoverished migrants would find themselves poorer in the rich regions. Mobility could play a positive economic role only if policies mandating an equal treatment of equals had been implemented.
After dating and discussing Buchanan's earliest views on federalism, we shed light on how he came to them. This is the second, and complementary, objective of this paper. We derive these lessons by studying the two essays in the context in which Buchanan wrote them. Let us recall that by the time he enrolled at Chicago at the end of World War II, Buchanan had studied with Carlton C. Sims at Middle Tennessee State Teachers College and with Charles P. White at the University of Tennessee in Knoxville. With Sims, he became acquainted with the question of the optimal size of counties, in particular in Tennessee.Footnote 3 With White, a specialist in taxation and federalism (see for example White, Reference White1931), and under his supervision, Buchanan had written a thesis for his MA degree on how to allocate the product of a tax among counties in Tennessee.Footnote 4 These questions had to do with institutions, public finance, and vertical relationships in a state, and they could be raised also at the federal level. But neither Sims nor White was of great help when it came to economics. Sims was a political scientist and, in Buchanan's own words, he “learned little or no ‘economics’ in [his] preferred definition during that Knoxville year” (Reference Buchanan2007: 69). Buchanan “learned economics” from his “Chicago teachers” (Reference Buchanan2007: 216). Who were they and what was their influence? We try to answer to this question, complementing what Buchanan wrote in his autobiography. We analyze the role that Theodore Schultz, D. Gale Johnson, Henry Simons, and Roy Blough played at the beginning of Buchanan's career.
Two economists deserve specific attention: Frank Knight and Henry Simons. Buchanan wrote the essays we discuss here after having taken, in the spring quarter of 1946, “Economics of Fiscal Policy” (EC 361) – taught by Simons – and “Price and Distribution Theory” (EC 301) in the summer quarter, taught by Knight.Footnote 5 Both Knight and Simons may thus have both directly and indirectly influenced Buchanan. All the more so as Simons was also influenced by Knight. However, in 1946 and 1947, as in Knoxville in 1940–1941, Buchanan was concerned with concrete problems in public finance. This was what Simons dealt with in the essays that he – or Blough, when he took over Simons's courses after the latter passed away – assigned his students.Footnote 6 Thus, what Simons was working on was more adapted to what Buchanan was looking for than what Knight discussed in his work and his lectures. It is not surprising that Simons influenced Buchanan in writing these essays more than Knight. More precisely, it is what Simons wrote on taxation and on fiscal justice rather than his work on the rules of the social game that Buchanan used in his two early papers. This is why we insist on Simons. We present some of Simons's ideas that Buchanan used in his essays and show how he used them.
2. Simons: fiscal justice, taxation and rules of the game
In his work, Simons in particular insisted on the fiscal relations between the different levels of governance in a federal state and, more specifically, on the difficulties that certain units of government had in meeting their responsibilities. The “incompatibility” (Reference Simons1935: 267) between “[t]he division of functions and expenditure responsibilities” (267) and “the division of revenue sources” (268), Simons noted, was a source of increasing concern for all “[s]tudents of government finance in the United States” (267). Simons made these remarks in a review of a book that would be important for Buchanan, The Principles and Problems of Federal Finance, written by Bhalchandra P. Adarkar and published in 1933.Footnote 7 Simons praised Adarkar's book despite its “shortcomings.” To him, “many passages and chapters … ought to be included in [the] assigned readings” in government finance courses (Reference Simons1935: 269).
The problem mentioned in the 1935 review of Adarkar's book was sufficiently important for Simons to mention it again in Personal Income Taxation: The Definition of Income as a Problem of Fiscal Policy (Reference Simons1938). In the United States, Simons wrote,
the kind of levies which represent the proper contributions of the state and local bodies to our total system of taxes are inadequate to their expenditure responsibilities. Many of them could not abandon their existing income taxes and death duties without serious disarrangement of their finances; and most states now rely largely upon undesirable revenue devices. (215)
These were the issues Sims and White had studied. Simons was thus discussing a point that could have reminded Buchanan of Sims's, White's or even his own concerns as well as the debates in the previous decade about state governance and the allocation of responsibilities between different levels of government.
In addition, Simons spent quite some time discussing the gasoline tax (Reference Simons1938: 38, 49, 205, 215) and other ad rem or impersonal taxes – which, one must not forget, had been the topic of Buchanan's essay for his master's degree. Now, to Simons, the gasoline tax was one of the “good ad rem levies” (30, 40), because it was one of the rare levies for which one could determine easily the benefits individuals receive from using goods financed by government. Or, in other words, this meant that one could apply the “doctrine of taxation according to benefit” (3), according to which taxes were “the prices against which people set the utilities of these [public] goods” (Reference Simons1937: 714). That is, “[e]ach person may be called upon, as in his dealings with private enterprise, to pay according as he receives” (Reference Simons1938: 3). Those taxes were useful and relevant, even though they were not based on an individual's “ability to pay” (5).
However, Simons claimed that, most of the time, those benefits could not be evaluated – “there are and can be no very satisfactory techniques either for determining total ‘benefit’ to property or for allocating the charges over particular parcels in accordance with relative benefit” (36). Hence, “the otherwise decisive case for benefit levies is seriously weakened” (37). Thus, concluded Simons, “it is one thing to say that levies should be allocated in accordance with benefit; and it is something else (apparently impossible) to specify in principle how this result may be achieved” (37; emphasis added). It is therefore not surprising that the benefit theory “has been repudiated as completely by students as by legislatures” (3) and that “this principle … is now of interest only for the history of doctrine” (3). It can be used for specific cases however not to explain “the allocation of the whole tax burden.” Thus, Simons was opposed to a benefit theory of taxation.Footnote 8 This led him to criticize Antonio de Viti de Marco, one of the Italian economists Buchanan would so much admire, for having “rigid[ly], if not explicit[ly]” (Reference Simons1937: 714) adopted it.Footnote 9 That was also why ad rem taxes could play an “important” (Reference Simons1938: 205) though only “subordinate” (205) role in a theory of taxation. When the benefit theory cannot be used as a general theory of taxation, and because ad rem taxes can be used only in some cases, other taxes have to be designed. How? To Simons, taxes must be “judged mainly in terms” (205) of two criteria – two criteria of the utmost importance for understanding the criteria that Buchanan himself used to evaluate taxes and a system of taxation. The first criterion was “their effects upon the degree of economic inequality” (205). Simons repeatedly emphasized this point, as it was noted by reviewers (Davisson, Reference Davisson1950; Sufrin, Reference Sufrin1950: 82; or later, Shaviro, Reference Shaviro2013: 13–14; Staudt, Reference Staudt1996: 652). He even noted in his Positive Program for Laissez-Faire how little inequality should be tolerated: “A substantial measure of inequality may be unavoidable or essential for motivation; but it should be recognized as evil and tolerated only so far as the dictates of expediency are clear” (Reference Simons1934: 12). And he devoted a large part of the introduction of Personal Income Taxation discussing different criteria of justice and defending the need to reduce inequality, for instance tellingly writing that “one may assert a substantially equalitarian position; or, at least, that there is a presumption in favor of equality and that the burden of proof rests with him who would depart from it” (Reference Simons1938: 17).
Unsurprisingly, Simons could not but view the reduction of inequality as one of the “three main objectives of liberal policy” (Reference Simons1936b: 72).Footnote 10 To reach that end, one should use taxation, and more specifically income taxation (Reference Simons1938: 41) More precisely, taxation had to be progressive. Certainly, it would affect efficiency and reduce incentives to adopt efficient behaviors, Simons was aware of the problem: “[i]t is reasonable to expect that every gain, through taxation, in better distribution will be accompanied by some loss in production” (19). But ethical objectives should always be combined with efficiency. Revenues should be maximized without forgetting “to secure an equitable, progressive distribution of tax burdens among individuals” (157). Simons devoted to this theme one part of the course that Buchanan took, using the first chapter of Personal Income and Taxation.Footnote 11
This leads to the second criterion that Simons insisted to use to judge taxes: just tax burdens “should bear similarly upon persons whom we regard as in substantially similar circumstances, and differently where circumstances differ” (30; see also 108, 148, 184, 205). Only one reviewer of Simons's book noted that, to Simons, “equity in matters fiscal obtains, if similarly circumstanced taxpayers bear substantially similar tax burdens over reasonable time intervals” (Wueller, Reference Wueller1938: 439). Buchanan did too: he would adopt the same equity criterion.
Another similarity was that Simons gave a lot of importance to the role of the state and to institutions. To Simons, economic problems were also political problems and vice versa. Government finance and, more broadly, the economy did not function independently from institutions, in an institutional vacuum. This perspective was close to Sims's and White's. Thus it certainly did not surprise or disturb Buchanan. But Simons went farther and, to a certain extent, deeper than they did – in particular, because of his references to the works of the founders of political economy. Simons insisted – this was why he found their works important – that Adam Smith and Jeremy Bentham were not economists only. They “stand out … as the great political philosophers of modern democracy” (Reference Simons1941: 213). And their views of democracy,
[t]heir special insight was that political and economic power must be widely dispersed and decentralized in a world that would be free; that economic control must, to that end, be largely divorced from the state and effected through a competitive process in which participants are relatively small and anonymous; and that the state must jealously guard its prerogatives of controlling relative prices (and wages), not for the purpose of exercising them directly itself but to prevent organized minorities from usurping and using the common interest.” (213–214)
This was put forward in the Positive Program for Laissez-Faire, and in many of his other works: governments have “unquestioned positive responsibilities … under the free-enterprise system” (Reference Simons1936a: 2). In other words he viewed it as “possible (and desirable) to give stability to a competitive, free-market, free-enterprise system without impairing its competitiveness, and without substituting political (monopolistic) for competition controls in the markets for particular goods and services” (Reference Simons1944b: 341).
These references to free enterprise, free competition and free markets are crucial. This is where Simons's position and Buchanan's pre-Chicago position diverge. Simons was really a pro-market economist, a defender of free markets and free competition. He also viewed the economic and political (institutional) aspects of problems as intertwined. To him, a free market system could not function without adequate political institutions. This implied that one could not defend freedom in the economy, a free market and free competition, without defending freedom in politics, that is, federalism. The alternative was simple: “we must choose between freer competition and increasing political control” (Simons, Reference Simons1936b: 68).
Simons's opposition to economic centralization, to which Buchanan had not been totally opposed until that date, concerned monopoly and pressure groups. Both aspects were connected. He targeted any large and powerful group in the economy.Footnote 12 They turned the economy into a “battleground” (Reference Simons1942a: 195), transforming economic activities into “warfare” (171), “organized economic warfare” (Reference Simons1944a: 4), or “a chaotic civil war of mass functional minorities” (Reference Simons1942b: 619), which was “like organized banditry” (Reference Simons1944a: 4), each group fighting against, or bargaining with, others “to advantage itself at the community's expense” (Reference Simons1942a: 195). Their activities – just “organized extortion” (Reference Simons1944a: 4) – usurped the sovereignty of legitimate institutions, threatened democratic institutions – American democracy – and led to dictatorship (Reference Simons1936b: 75) and authoritarianism (Reference Simons1941, Reference Simons1944a). To Simons, these monopolies/pressure groups usurped the sovereignty of legitimate institutions and even dominated the state (Reference Simons1948, 43). They threatened American democracy.
If economic centralization threatened freedom and democracy, the reverse was also true: political centralization threatened a free market system. To Simons, centralization was acceptable for certain goods and services (such as railroads), but only to a certain point, beyond which any “extension of federal responsibilities” was “unhealthy” (Reference Simons1935: 268). And elected officials and parliament members were incapable of resisting the pressures of lobbyists. The regulation of markets and interference by governments in economic activities ineluctably reinforced the power of monopolies/pressure groups and generated more regulation. Hence, a defense of free competition could not but imply a defense and desire to preserve federalism as it existed in the United States in particular – “[t]he maintenance of a vital sort of state and local government, and the reservation of large freedom and large responsibilities to the smaller jurisdictions, are indispensable for the preservation of representative political institutions” (Simons, Reference Simons1938: 215). Free markets, free competition and federalism were complementary and linked together.
Simons thus defended a position different from the one Buchanan had adopted in Reference Simons1941 – at that time, Buchanan had claimed that the gasoline tax should be managed centrally by a non-political body according to rules that should be changed every year or every two years. For his part, Simons criticized the idea that a society could be managed by technical/administrative agencies or by technical administrators (for instance, Reference Simons1942b). A free society, a free market economy, could not be “managed” as a firm. That resulted from his views on monopolies and centralization. Management, on such a large scale, meant interference in the economy. It implied a discretionary use of power that would be authoritarian and also inefficient. It would lead to instability and “extreme uncertainty” (Reference Simons1936a: 3), in the face of which no “enterprise system” could function.
Thus Simons pleaded for “definite rules”: “[t]he liberal creed demands the organization of our economic life largely through individual participation in a game with definite rules” (Reference Simons1936a: 1, original emphasis). The latter should be as general as possible – “it is only in terms of general rules or principles that democracy, which is government by free, intelligent discussion, can function tolerably or endure” (Reference Simons1944a: 2). They should, in other words, resemble constitutional rules. Precisely, Simons referred to the “‘constitutional structure’ under which free-enterprise economy and representative government can function” (Reference Simons1936a: 15). That echoed, and might have reminded Buchanan of, the importance White gave to the constitution from a public finance perspective. There was a difference: Simons did not view the constitution as a means for the government to increase revenues. A general structure, the constitution was supposed to be “the framework of our existing economic system” (Reference Simons1938: 2), which he significantly called the “rules of the game” (Reference Simons1938: 2). And he insisted on the “importance” for “a free-enterprise system that there should be considerable stability in the ‘rules of the game’” (Reference Simons1938: 32–33).
Indeed, Simons offered a radically different perspective with which to study the questions that Buchanan had been confronted with or had already studied before arriving at Chicago. As Buchanan learned when he took Knight's course just after Simons's, Knight had a very similar perspective on the economy. Knight's teaching thus complemented Simons's.
3. Buchanan's first term paper: ethics, mobility and the efficiency of federalism
Buchanan wrote “‘Federalism’: One Barrier to Labor Mobility” (Reference Buchanan1946b) for “Resource Administration and Policy” (EC 355), a course usually taught by Theodore W. Schultz, the 1979 Nobel Prize laureate in economics, and one that Buchanan did not take, as he recalled, just after his arrival at the University of Chicago (Reference Buchanan2007: 70) but in the fall Quarter of 1946 (see University of Chicago Annual Register, 1945–1946: 224). The head of the economics department – he had been since the beginning of the year, after Simeon Leland left for Northwestern University – Schultz was also involved in the nascent “Free Market Study” project.Footnote 13 Whether or not it was the reason, the fact is that he shared the teaching with one of his former students, D. Gale Johnson.Footnote 14 The latter, in particular, devoted two class periods to “Labor Mobility” (November 11, 1946) and “Migration Policies” (November 13, 1946). Also, among the works students were required to read for the course, one finds an article by Johnson on “Mobility of the Human Resource in Agriculture” – based on a section of his dissertation entitled “[a]llocation of resources between agriculture and the rest of the economy” (Reference Johnson1945: 186–192), and still unpublished when Buchanan took the course.Footnote 15 Thus, in some of the classes he taught, Johnson put the focus on labor mobility in agriculture and also between occupations. Many of the assignments given to students bore on this theme. This was crucial for Buchanan.
As was the emphasis that Schultz and Johnson put on dysfunctions in the agricultural labor market. On October 20, in a meeting of the class led that day by Schultz, students were asked to study “Occupational Adjustments and Maldistributions,” the sixth chapter of Colin Clark's Reference Clark1940 work The Conditions of Economic Progress (Johnson, Reference Johnson2008: 234). In the next class meeting, Schultz explained to students that “[m]al-allocation of resources occur (sic) in the long run – despite great pressures for shifting the shifts do not occur” (236) – and warned them that “[m]al-adjustments of resources will be the primary subject for the next few weeks” (237).
And then, on November 23, now discussing “short run considerations,” Schultz spoke of “mal-allocation” or “mal-adjustment” of resources and mentioned the role of “political power.” He noted: “commercial farmers and labor groups [are] concerned with price and wage rates not long time allocation of productive resources. Products become priced so as to bring about mal-allocation of resource [sic]” (253).
The problems linked to occupational mobility that Schultz and Johnson were talking about were of a different nature from the problems Buchanan had faced in Tennessee, in particular when he was going from home to college every day in 1939 and 1940. Yet the issues were connected: an inadequate transportation infrastructure could impair occupational mobility and generate an inefficient allocation of resources. Also, because he had lived in a state that was becoming more industrial and less agricultural, Buchanan could not have ignored the growing issue of occupational mobility. Thus, Schultz and Johnson gave Buchanan the tools to analyze a problem that had been important for him and was important for his state. At least, these reasons may explain why he chose to devote his term paper to “the mobility of labor from agricultural areas into the non-agricultural” (Reference Buchanan1946b: 2), “the migration out of agriculture into the non-agricultural sector” (2). It was Buchanan's first analysis of a phenomenon that remained crucial for him for decades (see Buchanan and Goetz, Reference Buchanan and Goetz1972).
Buchanan thus analyzed mobility and migration in terms of obstacles, “barriers and restrictions,” those that “serve as forces deterring movement toward this [optimal] allocation of the human resource in our national economy” (Reference Buchanan1946b: 2). More precisely, rather than analyzing the economic factors that could prevent mobility, Buchanan focused on “the institutional barriers and restraints within the economy which restricts mobility” (1) and, even more specifically, on one of those barriers: federalism. The title of his essay indeed leaves no doubt as to what he was arguing.
The first section of the essay explains how important is the adequacy of political institutions to the economic system. Such was the case at the end of the 18th century, when political decisions were almost always made by local units of governance and at the same time the economic system was also organized around local interactions. Then, progressively, infrastructures developed and the economy expanded and became national. At the same time, the sphere of government activities also expanded. It was no longer only a “protective” state but also took on the responsibility of “encompassing redress of inequalities though collective distribution of services” (5). However, the political structure did not change; it remained federal, with its “three levels, national, state and local, each of these levels performing certain of these social functions” (5). Now, Buchanan went on, at the lower level – that is, for local or subordinate units – the “geographical boundaries [do not] coincide with those of the economic market” (5). Thus, the immediate consequence of this was that “[e]ach unit is limited in its performance by the resources at its command” (5).
This is precisely the principle on which federalism works. But, like many others in those years, Buchanan viewed it as a problem. He stressed that “many of the state governments operate on a scale less than that of optimum size for administrative efficiency” (13), which was then put forward, including by Sims and White, precisely to justify county consolidation and functional integration in the 1930s (see Marciano, Reference Marciano2019). This had been present in Buchanan's Reference Buchanan1941 essay. It had also been identified as a central issue for fiscal policy in federal regimes, by Simons, Leland and even Blough, particularly in a Reference Simons1935 article that Buchanan listed in the bibliographical references without citing in the document.
The consequence, Buchanan noted, was that “there is a wide divergence among these units in the amount of resources contained” (5) and that “the value of public services performed collectively varies extremely among different horizontal units with the political hierarchy” (5). Thus, people living in poor states did not have access to the same amount of public goods and services as people living in wealthy states. That was a problem of justice, a form of inequality. Buchanan referred to the “very unequal provision of these [public] services (in value term per capita) among these horizontal units” (12). That inequality generated a second form of inequality: not only were these people poorer but they were also forced to bear unequal tax burdens – which has a Simonsian flavor, even if Buchanan made no reference to Simons.
This was important for him because of the “tremendous divergence” (7) that exists between horizontal units. As a consequence, the poorest states have a lower “tax paying ability” and therefore provide fewer public goods and services than richer states (4–7). It echoed what he had suggested in 1941: individuals were treated differently depending on where they live. Buchanan did not elaborate on that. He simply noted it, quoting Alvin Hansen and Harvey Perloff, for whom “the accident of place of residence of the American citizen determines in large measure the adequacy of the educational, health and recreational facilities with which he is provided” (in Buchanan, Reference Buchanan1946b: 7).
These different fiscal capacities, and the corresponding lower amount of goods and services supplied, have consequences for mobility. The point had already been made by various authors. Buchanan cited Clark, whom he had read for the course for which he was writing this paper, but did not cite Blough even though he listed his 1935 article in the bibliography. Thus, Buchanan seems already to have been acquainted with the work of the one who supervised his dissertation, even though it is hard to tell how much.
Buchanan combined these analyses to make his point. This led him to put forward five reasons that, according to him, explain why poor regions were disadvantaged compared to rich ones. First, because they benefited from fewer collective social services, individuals from poor states are physically and psychologically less mobile than those who live in wealthier states. Second, when they actually migrate from one region to another, they end up in lower paying jobs – “migrants tend almost exclusively to enter the unskilled labor ranks” (8). Now, if more social services were provided in poor regions, workers could migrate and claim better paying jobs. Buchanan accepted the idea that the relative poverty of certain areas contributes to a misallocation of human resources. This effect, and this is the third problem, is reinforced by the fact that population increases more in the poorer than in the other regions and, since people are poor and have no chance of finding better paying jobs (since collective services are not sufficient), these people remain poor. The situation worsens as time goes by because these individuals cannot find jobs even in nearby areas. Indeed, tax burdens – higher in poorer than in richer regions – reduce “the inflow of capital investment” (10). As a consequence, we have the fourth problem: the demand for labor is also reduced. Finally, a fifth factor limits the demand for labor and the inflow of investment: minimum wages. As legitimate as this kind of policy can be to avoid labor exploitation in the whole economy, it cannot but create problems in poor regions. The reason is simple: “due to the fact that unequal resource distribution has created long run divergencies in investment in the human resources, productivity is lower in the economically poorer regions” (11). Hence, if obliged to pay the same wage rates as richer states, employers in poorer states would hire fewer workers, where labor productivity is lower. This last point was related to the regulations Roosevelt had tried to implement with the New Deal policies.
The conclusion Buchanan derived was clear and straightforward:
the existence of subordinate political units within the national economy including within their boundaries [with] very different taxable capacities (income plus capital) but attempting to perform services comparable with each other in the nature of investment in the human resource … has thereby served as a significant force limiting and directing the mobility of labor off the agricultural into the non-agricultural sphere. (12)
Thus, federalism prevents an efficient allocation of human resources: it “is one barrier to the attainment of long run equilibrium of the national labor force.” (3)
To solve the problem, the first and easiest possibility would be to abandon federalism and to centralize all the functions performed by the different levels of government. Indeed, he noted, federalism “seems to have few, if any, economic advantages, and many heavy and forceful disadvantages” (15). Centralization would be more efficient and more equitable: “[c]omplete unification and integration on a functional efficiency basis would not only distribute public service equitably, but would also be technologically more efficient” (13). This was the type of solution Sims, White, Leland and even, to a certain extent, Buchanan himself had defended in his 1941 essay for reasons of efficiency. Buchanan referred to Leland as “a strong proponent of such a unification” (13).Footnote 16 But that was not what Simons had argued. Simons rejected centralization for political reasons. This is where Buchanan joined Simons in his opposition to centralization and defense of federalism. His arguments were close to those of Simons.
First, he claimed that institutions cannot be viewed abstractedly, that is, independently from the society they organize. From this perspective, the advantages of a federal structure “are more than theoretical” (12). They result from the fact that federalism is accepted by most Americans – it is “a commonly held valuation by the American people” (12) and “is securely rooted in American ideas” (13). Hence, it was too late to change so radically the political institutions of the United States. Thus, albeit implicitly, Buchanan was defending federalism because it meant defending American democratic institutions – echoing what Simons had written. He explained that, even if institutions were relative to a country or a society, they progressively acquired an absolute value. After two and a half centuries, in the United States, federalism “must be accepted as a value in itself” (13) and therefore “must be preserved in principle” (14; emphasis added) – here, Buchanan's claim echoed Knight's about the value of democracy and freedom. And it was again Simons who resonated in the words Buchanan used to explain that it was “the fear of tyrannical usurpation of power” (12) that had “led to the establishment of the federal government in this country” (12). The raison d’être of federalism lies in its political advantages, namely, “the protection against overcentralization and the value of subordinate units for experimental development” (14). This was precisely why federalism could not be abandoned.
This did not mean that nothing should change. Fiscal reform was necessary. Leland, Simons, and others had made the claim. Buchanan agreed. He added that fiscal reform should have to guarantee one of the most important features of federalism, state sovereignty – “[t]he sine qua non of fiscal reform is the maintenance of state sovereignty” (14). This could be achieved by using the same devices that were already used and that involved the intervention of the federal government. Thus, Buchanan agreed that funds should be distributed by the central government to the subordinate units along the lines of the grant-in-aid device.
However, Buchanan suggested a change to the grant-in-aid system as it existed in the US at that time. To him, the system was flawed because it “allow[ed] the rich units to profit at the expense of the poorer” (17) and also because “the grants have been restrictively earmarked” (17). As a consequence, this led to “local expenditures not necessarily economically justifiable” (17). The new system should be adapted to a peculiar objective: the new grant-inaid device should now satisfy a “criterion of equity … that individuals in similar circumstances throughout the national economy should be treated similarly” (18). In other words, he defended the idea of using grants to promote an “[e]qual treatment of equals” (18) because that was “[t]he only criterion of an equitable distribution of central funds to subordinate units” (18) – which means that Buchanan clearly and explicitly adopted an individualistic perspective and which echoes Simons's view, noted above, that tax burdens should bear similarly upon persons whom we regard as in substantially similar circumstances. Thus, Buchanan started where Simons had started and ended up with a defense of egalitarianism similar to that of Simons.
Buchanan did not doubt that this criterion “is acceptable to all” (Reference Buchanan1946b: 18). It did not seem to him to be controversial, at least in terms of welfare – “[i]t does not hold that the rich should be taxed to support the poor” (18). It has also the advantage of being independent from any practical mechanisms of implementation – it does “not require to set up a basis on which funds should be distributed” (18). In addition, it does not require “that tax rates be equivalent in all subordinate units” (19). This is crucial: the criterion he put forward only requires that
an individual's residuum (plus or minus) of collectively provided governmental services available to him (federal, state, local) minus his tax payments (federal, state, local) should be the equivalent of that of any other individual residing anywhere else in the economy possessing equal income and living in equal economic circumstances. (18–19; emphasis in original)
Admittedly, that was an “ideal” that could not be reached because of “cultural and historical patterns, and administrative difficulties” (21–22). But the equity criterion was nonetheless important because it “does point a goal toward (sic) which we should strive, a way of action” (20). And that deserved to be studied more. Buchanan would use it in his first published articles in the early to mid-1950s. He also extended it in his next essay.
4. Buchanan's second-term paper: ethics and the inefficiency of federalism, again
In 1947, Buchanan took a course that proved to be much more important for his career: EC 362, “State and Local Taxation.” Leland had once taught the course but, as mentioned above, by the winter of the academic year 1947–1948 he no longer did. The instructor was now Blough, who had been hired in the fall of 1946 to replace Simons after the latter passed away.Footnote 17 They knew each other and Simons appreciated his competences. He viewed Blough as the “best available tax specialist,” one of these “specialists who are not also broadly competent economists and not useful outside their specialties” – and this was different from what Simons and Buchanan were: economists with an interest outside their specialties and even outside the discipline itself. A former director of tax research at the Treasury Department and assistant to the treasury secretary from 1938 to 1946, Blough had also written academic articles on public finance and taxation. And his and Simons's views “on theory and policy … were much closer … than their pedigrees might imply” (Johnson, Reference Johnson2014: 2). Yet both had different views about free markets and decentralization. Blough was “another New Dealer from the University of Chicago” (Johnson, Reference Johnson2018c: 103). However, this may not have mattered so much, and Blough was technically competent to judge Buchanan's work and possibly guide him, particularly if one focuses on the questions that were important to Buchanan.
To start with, Blough was also – even if less – convinced that taxation and fiscal policy were institutional questions. In a 1944 article, Blough insisted that taxation was not simply an economic matter but also depended on the social and political objectives that a society wanted to achieve:
TAXATION is an instrument of organized society. Itself a social institution or group of institutions in the broad sense, taxation exists for the service and promotion of other institutions; in and of itself it produces no personal or social utility. Accordingly, the system of taxation rests on the political, economic, and social structure and should be fitted to that structure and adapted to achieving the ends and objectives of the society. (Reference Blough1944b: 22)
Among those ends it should aim at, social justice was one of the most important. But this was not a goal that taxation seemed to be really fit to reach: “[t]he removal of social injustices and the achievement of social justice are a very heavy load to place on the tax system” (Reference Blough1944a: 7). Certainly, to Blough, taxes should be just (6), but he introduced a distinction between “social justice” and “justice in taxation” that Simons or Buchanan had not made.
Regarding “just taxes,” Blough also referred to a concept to which Simons had referred and that Buchanan had already endorsed: “If John Brown and John Smith both are in the same economic situation, tax justice requires that they be treated alike” (6). But he immediately qualified this statement. Satisfying this objective did not really allow a state to achieve justice but rather equity, which he viewed as a narrower form of justice – “[t]he equitable treatment of persons who are substantially alike in their economic situations is often referred to as tax equity as distinguished from the larger concept of tax justice” (6). Also, Blough did not seem to distinguish conceptually an “equal treatment of persons who are substantially alike in their economic situations” from another criterion, namely “ability to pay” – although he distinguished them formally. He thus wrote that “[t]ax equity consists in recognizing these inconsequential differences for what they really are and adjusting taxes so that they fall with equal weight on people with equal ability to pay taxes” (6). That might be explained by his defiance – similar to that of Simons – of a benefit theory of taxation (Reference Simons1944a: 13; 23).
Thus Blough's work showed that scholars with different views on the economy could converge toward the same objective and the same or a similar definition of justice in taxation. Complementarily, it also evidenced that an equal treatment for equals was not so universally accepted a criterion of justice as Buchanan thought it was. Buchanan had to justify it before he could use it. More precisely, he had to show how his criterion of justice was compatible with benefits, and hence had to clarify the role of benefits in a theory of taxation.
Whether or not Buchanan was influenced by Blough in those matters remains a matter of speculation. It is nonetheless true that Buchanan raised these very questions in the term paper he wrote for Blough's course, “A Theory of Financial Balance in a Federal State” (Reference Buchanan1947).
This essay grew out of and expanded the one written in 1946. Even if the perspective was more theoretical and more abstract than in the 1946 paper, and although Buchanan no longer focused on a specific economic problem – occupational mobility in agriculture – the starting point was the same in 1947 as in 1946, as was his description of the problem that only federal states have to face. Now, a major difference, Buchanan named the problem, speaking of “the dilemma of federalism” (Reference Buchanan1947: 1). Buchanan was actually not the first to use the word “dilemma” to characterize the financial problems of federal regimes in this way (Warner, Reference Warner1933: 117; Corry, Reference Corry1941; Ross, Reference Ross1943: 888, 889; Maxwell, Reference Maxwell1946: 39). It was James A. Maxwell and his work The Fiscal Impact of Federalism in the United States that Buchanan cited. In his book, Maxwell described how federalism in the United States had led to a progressive growth of government responsibilities and to centralization and to a disparity between the fiscal powers of the federal government and the tasks local units still had to perform. Maxwell nonetheless defended federalism against centralization – “American federalism has fiscal problems for which outright centralization is no cure.” The solution to the dilemma should consist in using a grant-in-aid system with conditional grants. Buchanan shared the diagnosis but disagreed with the solution.
Before Maxwell, J. A. Corry, a Canadian political scientist and specialist in administrative law, had used the expression “federal dilemma” as the title of one of his articles (Reference Corry1941).Footnote 18 He had characterized the situation as Buchanan would: there were “centralizing tendencies” – political and economic centralization “ha[d] been going on without significant interruption for over fifty years” (Reference Corry1941: 216) – that “weakened … the foundations of federalism” (217). To solve the problem, one had first to figure out whether “the existing … states still represent genuine unities (sic) of interest and that particularism in one or other of its important manifestations will therefore rally around them rather than around some other territorial, occupational or class grouping” (228). The solution was institutional. Buchanan referred to Corry in his dissertation, but not in his 1947 essay. Yet his analysis was similar to Corry's and his work can be viewed as trying to deal with Corry's last remark about how to solve the dilemma.
Before reaching this point, Buchanan had to explain in what the dilemma consists. Buchanan repeated his claim from 1946: the “divergence [between] the boundaries of the political and the economic entities” (Reference Buchanan1947: 10) was a consequence of the expansion of the economy, of the size of the market – “the economic sphere has continually expanded … the boundaries of the economic market have expanded at the rate determined by the rate of expansion of technological progress” (8–9). But, political centralization remained limited compared to economic centralization. Local units still had economic functions to perform – their sovereignty remained at least partially intact. Now, he wrote, exactly as in 1946, local units have “different quantities of economic or fiscal resources” (10), and therefore, some of them are “limited in the performance of [their] functions” (10). They do not have enough means to finance the public goods they are supposed to provide. Hence, he wrote as he had already written in 1946, that in a federation there exists “a wide variation in the level of welfare and social services provided to individuals in the different subordinate units, in rates of taxes levied, and in number and intensity of function performed” (10).
Thus, in a federal state, individuals do not benefit from the same quantity of public goods and services because of their “geographical location” (18). Federalism lacks geographical neutrality (24; 32; 45). The victims of “the accident of place of residence” (32), individuals are treated differently because of where they live, even when they are in similar situations. Hence, federalism violates the principle of horizontal equity that Buchanan had already mentioned and defended: “the basic equity principle of equal treatment for equals” (Reference Buchanan1947: 20) or for “similarly situated individuals” (35). This principle, Buchanan noted again, was “universally accepted” (18, 31). At least it was accepted by Henry Sidgwick, whom he quoted from the very same book Simons had reviewed, and he also suggested that students of public finance read Adarkar's The Principles and Problems of Federal Finance. This was a genuine innovation compared to the literature on the dilemma of federalism. Buchanan was the first to approach the dilemma in ethical terms.
The negative effects of geographical location, and the problem of injustice, would disappear if individuals were mobile and could migrate from a poor to a rich state. But this was not the case. As in his 1946 paper, he repeated that individuals are not all equally mobile and are less mobile in the poorer than in the richer states. Not that their tastes or preferences are different. The reason was the “lack of knowledge of opportunity, low productivity, illiteracy, low health standards, etc.” (44) that comes from “the deficient provision of the social services, especially education and health” (44). Even the individuals who move to another area, Buchanan then added as in 1946, will not receive higher wages. They “are forced into the unskilled ranks for the same reason” (44). In other words, mobility could not be viewed – in contrast to what Tiebout would write in “A Pure Theory of Local Expenditures” (Reference Tiebout1956) – as a solution to the existence of economic differences between local units. The lack of mobility was a consequence of the differences between the states and of the related fiscal imbalance – a consequence of federalism, actually: “[p]eople are restricted from moving from area to area to equalize incomes by the very existence of the institutional structure” (44).
More broadly, it seemed obvious to Buchanan that federalism was not an efficient institutional structure – “excellent arguments can be made to prove that the effects of the federal structure with its uncorrected varying capacities of its subordinate units to provide social services are directly to restrict optimum resource allocation” (44). Other institutional arrangements had to be envisaged. The most obvious alternative – the “easy solution to the problem” (11) –would be to centralize economic and political functions. This would imply “a complete fiscal integration” (11) and “a completely unified government” (11) that would lead the boundaries of the economic entities – the national market – to coincide with those of the political unit – the nation.
Leland had defended that solution in his work. And, as in 1946, Buchanan cited him. He contrasted Leland's views on centralization with his own on decentralization and the need to preserve federalism. Even though Buchanan praised Leland's proposals – they were “far superior to the weak proposals of those who would retain the federal structure in this country” (11) – he also rejected them and again rejected centralization as a solution to the dilemma. A solution to the dilemma of federalism “must begin with the semi-fiscal independence of the states as a datum” (12). Federalism should thus be preserved despite its disadvantages over centralized political regimes. Because of its intrinsic value: “the principle of federalism is desirable in itself” (13). At least for the USA. Indeed, as in 1946, Buchanan wrote that the “fundamental advantages to a federal structure … are now more than theoretical; they are ingrained in the thinking of the American people, and a federal political structure represents a commonly held value premise” (12). In a different country, with a different tradition and history, the perspective would be different – “should the valuations of society change, the alternative of fiscal unification and integration is always available” (13).
Buchanan did not only reject the (easy) solutions that would consist in abandoning federalism. He also criticized and rejected the proposals that had been so far made to solve the dilemma, and to reduce inequalities between different regions. All these proposals were “almost without exception based upon the idea that any solution to the dilemma must, in effect, be a compromise of sorts” (Reference Buchanan1947: 2). The problem did not lie in compromise itself – “all economic institutions have, in a sense, evolved through a compromise” (2) – but how it was adopted. The “policy direction taken in this field” (4) was according to Buchanan too pragmatic. It “follow[s] no concrete pattern” (4), “consist[s] purely of action in response to pressing and urgent situations … [and] amount[s] to particular adjustment to particular circumstances” (4). That was precisely what he had noted in his 1941 analysis of the administration of the gasoline tax funds – political actions made in response to the demands of voters. No one – in particular the “experts” (3) – seemed to have envisaged the “problem in its entirety and to have established basic principles upon which really significant practical measures could be constructed” (3). A theory was missing.Footnote 19
It was therefore not surprising that no progress was made. As Buchanan explained, to solve the dilemma, one needed “principles,” or a “theory.” This was what he intended to provide and this is where the 1947 paper differs from the 1946 one: it is aimed at providing a theory of financial balance in a federal state. An ambitious goal, since no such theory exists – “[t]hose who would retain the federal systems have done no such thing” (11–12) – and the existing literature offered little help to guide him. He had found only one reference that adopted an approach similar to the one he wanted to adopt: again, Adarkar's book. Its merit was that the problem of federal finance, of financial imbalance in a federal state was envisaged from “an integrated” perspective, “largely abstracted from practical political considerations,” by “separat[ing] out the basic issues in the federal financial problem without confusing them in the maze of multifarious interconnecting relationships confronted in the second order considerations of this field” (3).
The solution Buchanan then proposed was that the central or federal government should proceed to “interarea transfers” from rich to poor units. It was not very original. Most federal countries used them and many public finance specialists, including the ones Buchanan cited or quoted – Adarkar (Reference Adarkar1933), Maxwell (Reference Maxwell1946) and Blough (Reference Blough1935) – defended them. Buchanan himself had already mentioned that possibility earlier. He gave more details, in particular regarding the three points around which most discussions revolved: the need to guarantee a minimum level of certain goods or services, the capacity of states to manage the transferred funds, and the form these transfers should take.
First, Buchanan did not oppose the idea that, when there existed “a strong enough national interest in providing a ‘minimum’ … level of particularistic services” (Reference Buchanan1947: 34–35), “the function should be taken over and assumed as a responsibility of the central unit, and equal provision of the service made throughout the economy” (36). In specific cases, states may indeed be obliged to use the funds transferred for specific purposes decided nationally. That would prevent some states from neglecting the provision of important services, such as health care and education. Otherwise, the transfers should be administered by the local units themselves. In other words, the federal government should transfer funds but not encroach on the state's prerogatives.
This led to the second point, the possibility that local public decisions may lead to misuse of the resources transferred by the federal government. Buchanan had already stressed the threat in his 1941 MA thesis. Buchanan repeated himself:
[t]here is no presumption that the unit can best use the funds granted in the performance of the service for which such funds are earmarked. The currently popular matching requirements of course doubly accentuate the distortion causing the unit to allocate not only the grant but also a portion of its locally collected funds to specific functions. (36)
This might be an argument to justify that “some conditions should be attached to [these grants] concerning ways in which the funds must be expended, the manner in which they are administered. This is necessary to prevent waste and graft” (37). But these conditions should be temporary. On the whole, local units should be trusted to use these funds optimally and, in particular, to meet the “‘adequate’ level” (34) of services in domains like education or health – “there seems a strong likelihood that the ‘minimum’ services needed would be provided” (35). This was why Buchanan suggested the abandonment of conditional – earmarked – grants. To him, grants should be “without strings attached” (35). In this manner, the “budgetary independence of the subsovereign unit” (35) would be preserved, which was “the fundamental principle of federalism” (35). This led him to plead for unconditional grants. This was the third problem that was discussed in the literature.
To Buchanan, earmarked conditional grants were problematic because they were “conditioned to serve specific needs” (34). The very notion of “need” was problematic. How to define it? It could not but be arbitrary. Even if the definition was guided by charity or Christian virtue, it would still be arbitrary (32). To Buchanan, grants were not justified because of a sacred – he used the German word heilige (32) – duty. They were made to restore equity by “providing geographical neutrality in the ability of the subordinate units to provide service standards” (45), that is:
[by] assuring to all people within the economy that the accident of place of residence did not determine their treatment by the political structure; that the accident of place of residence did not determine the amount of social services available nor the contribution exacted.” (32)
Thanks to these transfers, individuals will be treated similarly whatever their place of residence. Equals will be treated equally. Indeed,
the net result of the application of the principle of “equal treatment for equals” in taxation would be that the political units as a coercive force “pressed” equally on similarly situated individuals in the economic structure if the tax burden of the similarly situated individuals were the same. (16–17)
That was enough to solve the problem: “[t]he use of this criterion will resolve completely the perplexing problem. It is as simple as that” (20). Simple, at least from the perspective Buchanan had adopted: a very basic numerical example involving two states, A and B, and three individuals split into two income groups.
Then, Buchanan proceeded by stressing the advantages of the principle. To begin with, it was widely applicable, since it was independent of any type of tax system – “proportional, regressive, degressive or progressive” (17) – and of the way burdens and benefits are distributed. Also, the criterion could be applied independently of the level of taxes (burdens) and benefits. Then, the criterion did not impose anything on the units of governance: “A subordinate unit could tax its people heavily or lightly according to its own free choice without affecting the results of the application of the criterion which had been set up” (23). What matters is that “the fiscal residua left to them, or taken from them (as the case may be) were equivalent” (20–21). In addition, the fiscal residuum is the difference “between the economic value of the burden imposed by ‘government’ on the one hand, and the economic value of the services rendered to the individual on the other” (21). Finally, “[i]f a subordinate unit so desires by its own choice to tax its citizenry heavily to provide greater benefits, the residuum of an individual in that unit will remain the same as prior to the tax” (22).
Buchanan dealt with another difficulty, the possibility of determining individual fiscal residua and evaluating the benefits individuals receive in terms of public goods and services, which he had addressed in his 1946 paper. Recall that Simons had insisted on the “enormous difficulties” and the lack of “satisfactory techniques” to measure benefits from taxes and this was the reason why he opposed the so-called “benefit theories” and ad rem taxes. Buchanan agreed with Simons, admitting “the obvious difficulties involved in measuring or estimating individual benefits derived from government” (15) and “that benefits (except in special cases) could not adequately be imputed to the individual as such” (15). These difficulties had led to “the overthrow of the benefit theory as a basis for distributing the tax load” (16). But, departing from Simons's position, Buchanan regretted that the difficulties had also led to the abandonment of the “benefit idea.” He was convinced that, even “if benefits could not be imputed to individuals differentially” (16), they could nonetheless be measured as “a mere per capita portion of the total expenditure made in the ‘general welfare’” (16).
Thus, Buchanan found it unfortunate that a benefit theory of taxation was abandoned because benefits could not be measured. He insisted: the measurement of benefits, first, was a practical and not a fundamental obstacle and, second, could be overcome. But he had not closed the gap between, on the one hand, the need for an equal treatment for equals and, on the other, benefits and a benefit theory of taxation. In other words, he had not explained how to use a benefit theory of taxation in a public finance approach with the objective of treating equals equally. It was not until the writing of his PhD dissertation that Buchanan showed how it could be done.
5. Conclusion
Buchanan's defense of federalism dates back from when he was a graduate student at the University of Chicago and enrolled in courses taught by Schultz and Johnson, and then Simons, Knight and Blough. By putting his graduate school essays in the context in which Buchanan wrote them, we shed light on the influence these professors had on Buchanan – and explain why Blough supervised his dissertation. It thus seems that Simons's influence on these essays was greater than Knight's. This would (at least partly) change. When he wrote his dissertation, Buchanan would analyze more deeply the theoretical foundations of his ideas. He would read Wicksell, adopt a benefit theory of taxation and depart from Simons. In the long term, Knight – and Wicksell – were important to Buchanan's intellectual development. It was not the case at the time.
Even the importance of ethics seems to owe more to Simons than to Knight – at least, the evidence points more at Simons than at Knight. Buchanan was interested in ethics from the perspective of fiscal justice, which was more specific than Knight's views on ethics. This is particularly interesting because it helps us to understand that Buchanan did not promote a vision of unrestricted competitive federalism. To him, a federal regime without an ethical rule that would allow equal treatment for equals would be unjust. He would repeat this idea later, not only demonstrating the importance of ethics but also the broader impossibility of achieving efficiency in the absence of ethical rules. To Buchanan, as for Simons and Knight, a free market economy was indeed a system with rules.
Thus, Buchanan defended federalism because it was a system that the Americans were used to; it was part of their culture. It also appears that, even if it remains implicit and very much indirect, Buchanan already had a view of federalism as a regime that allows a state to experiment with solutions. One should be careful and not over-interpret what he then wrote. It was very early in his career but this cannot but make us think of Vincent Ostrom's (Reference Ostrom1987) theory of the “compound republic” and even Riker's (Reference Riker1964) “rational choice” bargaining view of federalism (for a recent comparison on Buchanan and Ostrom, see Bish, Reference Bish and Wagner2019).
Complementarily, Buchanan's view can be seen as anticipating very recent studies about how the experimental aspect of fiscal federalism (a relatively neglected theory) is tied to unconditional grants (Garzarelli and Keeton, Reference Garzarelli and Keeton2018). Indeed, we can think of Buchanan as anticipating much later views about the superior role of unconditional grants to preserve the nature and spirit of fiscal federalism (see for example Breton's general defense of unconditional grants for federal competition (Reference Breton1987: 312) and for genuine federal freedom (Reference Breton1996: 258).
Acknowledgements
I thank Peter J. Boettke, Paul Dudenhefer, Andrew Farrant, Giampaolo Garzarelli, Steven G. Medema, Vlad Tarko four anonymous referees and the editor of the journal for very helpful comments and suggestions.
