Two recent books help us to think through these issues: Democracy for Realists by political scientists Christopher Achen of Princeton and Larry Bartels of Vanderbilt, and Escape from Democracy by economists David Levy of George Mason and Sandra Peart of the University of Richmond.

Achen and Bartels remind us that the Founders were suspicious of democracy and instituted a republic with many checks on democratic controls, such as a Constitution including a Bill of Rights, the Electoral College, and two houses of Congress. Divided and limited government were features of the republic, not bugs.

Unfortunately, those institutions have proven less hearty than the Founders had envisioned. James Madison, one of the major proponents of divided and limited government, would be distressed at how the U.S. government and popular ideology have become democratic. He would not be alone. “Even Thomas Jefferson, often remembered as a dedicated democrat in a republican age, was anxious to limit the influence of the urban masses,” note Achen and Bartels.

Danger of experts / Federal bureaucratic power has grown, especially since the Progressive Era. The power of experts in and on government is the topic of Levy and Peart’s Escape from Democracy. It will be convenient to review this book first.

Levy and Peart argue that we should not escape democracy by accepting the rule of experts. They focus on economic experts, but their arguments apply to all experts who advise government. Following Chicago economist Frank Knight, they think of democracy as “government by discussion.” In this tradition, exemplified by such economists as Adam Smith, John Stuart Mill, Lionel Robbins, and Milton Friedman, “the good society is one that governs itself by means of an emergent consensus among points of view.”

Escape from Democracy claims that the discussion tradition “largely disappeared from the economics literature with the advent of new welfare economics” in the 1930s and its offshoot of cost‐​benefit analysis. Under this new school of thought, the goals of public policy are considered to be exogenously set by the citizens, while the experts analytically decide the means to accomplish those goals. Levy and Peart reject this notion. Discussion should continue about the means as well as about possible modifications of the goals. They reproduce a beautiful sentence from Knight:

In contrast with natural objects—even the higher animals—man is unique in that he is dissatisfied with himself; he is the discontented animal, the romantic, argumentative, aspiring animal.

The two economists insist that popular knowledge incorporates more wisdom than it is credited with. Their stance is contrary to that of Georgetown political philosopher Jason Brennan in his Against Democracy (Princeton University Press, 2016; see “Power to the Knowers!” Spring 2017), as well as in many ways to that of Achen and Bartels, as we will see. While the experts’ input is required, it should not be blindly deferred to because it is biased: the experts have their own private motivations, self‐​interest, and incentives.

Two chapters of Escape from Democracy analyze two episodes where experts seriously erred: eugenics in the first part of the 20th century (see “Progressivism’s Tainted Label,” Summer 2016), and the gross overestimation of Soviet economic growth by economists—including Nobel prizewinner Paul Samuelson—from the 1960s to the 1980s. For several years, Samuelson attributed to bad weather the languishing of the Soviet economy. During the Progressive Era, virtually all experts favored eugenics, and eugenic sterilization lasted until 1974 in North Carolina. We might add that today’s public health movement shows a similar authoritarian bias. (See “The Dangers of ‘Public Health,’ ” Fall 2015.)

Levy and Peart favor institutions that minimize the biases of experts. “The goal is to generate plenty of unbiased expert judgments, for instance, by making transparency incentive‐​compatible or, if that is unattainable, by making nontransparencies transparent.” In other words, experts’ incentives must be structured away from biases or at least their biases must be transparent, like they are in legal litigation. The authors emphasize the importance of codes of professional ethics and the disclosure of conflicting interests (in scholarly articles, for example). They also present one “radical proposal”: to have regulations approved by juries, an institution the authors see as “the paradigm of democracy.” With the help of a sophisticated statistical argument, they show that “something akin to jury trials might be a viable means to obtain the benefits of expertise in the regulatory setting.”

Levy and Pert are (brilliant) experts in the history of economic thought and their book is replete with original insights. They admit that experts, including themselves, have their own incentives, which may conflict with the search for truth. Incentives are the bread and butter of economic analysis, and we should “apply the tools of economics to the economists who use them” (emphasis in original).

Escape from Democracy is a learned, fascinating, and wide‐​ranging book. It goes from alchemy to statistical analysis of jury‐​type decisions. It raises many deep questions and proposes original approaches.

Perhaps its arguments can be extended or improved in a few directions. Levy and Pert write about “group (societal) goals,” but it is unclear what those goals are. They also talk about “society’s goals,” and even slip into “we as a people,” but I will blame their editor for the latter. Are these group goals some sort of collective goals, or are they instead the “shared goals” of individuals? How can they be aggregated? The authors do suggest in passing that these “goals” are Hayekian‐​type conventions, but their thesis would benefit from hashing through these issues.

I think that by rejecting welfare economics, the authors of Escape from Democracy have deprived themselves of a useful way to think about these issues. The first theorists of the “new welfare economics” were no doubt mistaken in thinking that, with the so‐​called Kaldor‐​Hicks criterion, public policies could be proposed without making value judgments. In two 1950s articles, Samuelson himself demonstrated that it is generally impossible to make welfare evaluations (about all individuals) without bringing in normative, extra‐​economic judgments about distribution. In fact, Samuelson supports Levy and Peart’s claim that economic experts cannot prove the superiority of some public policies over others.

It is not only Kenneth Arrow who, as Levy and Peart write, “called into question the soundness of the new welfare enterprise.” With Samuelson, welfare economics itself finally proved what it had set out to disprove—that is, the methodological capacity of economic experts to dictate public policy. The fact that Levy and Peart mention Samuelson as part of the discussion tradition suggests that they are open to putting some water in their anti‐​welfare‐​economics wine.

There is also a risk of exaggerating the import and justification of democratic discussion. Meaningful democratic discussion cannot cover everything and it should not try to. To the extent that it can result in coercive decisions, it should not encroach on private domains. (More on this later.)

I suspect that Levy and Peart would accept some of my objections—at least the last one, concerning the necessity to restrict the scope of democratic discussion. Achen and Bartels, on the other hand, would likely reject that objection.

Very different book / Democracy for Realists is a very different book, if only methodologically. It astutely marshals much empirical evidence and presents a host of interesting statistical analyses about what voters think and how democracy works. It is less rooted (if at all) in the school of rational choice and criticizes “economistic thinking.”

The two political scientists set out to “document the gap between democratic ideals and realities,” to show how the populist or folk theory of democracy is untenable and, in the second part of the book, to propose a new justification for democracy.

According to the folk theory of democracy, the voters “have preferences about what their government should do,” and they “choose leaders who will do those things” or “enact their preferences directly in referendums.” In this perspective, voters really rule, as opposed to Madison’s “trustee model” where they elect a body of wise citizens to realize the public good.

Achen and Bartels deploy a trove of data and statistical analysis to show that the folk theory of democracy makes no sense. Elections don’t represent “the will of the people” because preferences cannot be aggregated in any coherent way. Voters are ignorant of facts pertaining to politics and they vote for or against parties and candidates, not on actual issues. At the end of 1996, for example, a majority of voters, both among Democrats and (even more) among Republican, thought that Bill Clinton had not significantly reduced the budget deficit, which had in fact been cut by 40% in three years—or 58% if we count the fiscal year at the end of which the survey was run. Politicians are not much constrained by voters’ opinions.

The least demanding version of folk democracy, “retrospective voting,” does not correspond to reality either. Retrospective voting occurs when voters at least know enough to “throw the bums out” when election time comes and they have not been served well. But in reality, voters engage in blind retrospection. Statistical analyses show that they punish the incumbents (of whichever party) only for the economic conditions obtaining in the last six months preceding the election, whether or not the politicians in power could have done anything to change those conditions. They punish incumbents for droughts and floods. In 1916, voters even punished Woodrow Wilson and his Democrats for shark attacks in New Jersey.

Despite such interesting analyses, Achen and Bartels suffer from their own blindness. They ignore the explanation of “rational ignorance,” the fact that a single voter has, for all practical purposes, a zero chance of changing the election result. Consequently, Achen and Bartels do not see how the rational voter votes his opinions, not his interests (except to the extent that he has rationalized his interests into his opinions or his whims).

More generally, the authors neglect the economic literature on voting. They cite George Mason law professor Ilya Somin’s Democracy and Political Ignorance (Stanford University Press, 2013), but not the work of his Mason economics colleague Bryan Caplan (The Myth of the Rational Voter, Princeton University Press, 2007) nor that of political philosophers Geoffrey Brennan and Loren Lomasky (Democracy and Decision, Cambridge University Press, 1997). Except for the old work of Brookings economist Anthony Downs, they neglect the whole public choice analysis of democracy. James Buchanan, who won the 1986 Nobel economics prize, is nowhere cited. They don’t mention and seem to ignore the economic analysis of collective action. They view the work of economists on democracy as “limited” and “naïve.”

Instead, they borrow from psychology and the soft field of sociology to develop a “group theory” of democracy. Individuals, they claim, are moved by “group loyalties and social identities.” Puppets of racial and other social groups, individuals just obey group norms. They don’t support political parties whose opinions match theirs, but instead take their opinions from the political parties (and other groups) they belong to.

There is something true in the idea that groups are important, and Democracy for Realists offers supporting statistics in the realm of politics. But standard economic theory explains most of that in terms of rational behavior; no need to invoke some soft‐​sociological theory of groups. Some of the gregarious behavior of individuals can also be explained by our tribal wiring inherited from evolution, but Achen and Bartels don’t even cite Friedrich Hayek, another Nobel prizewinner, who has studied the implications of this fact for modern societies. Hayek also explained how methodological individualism is useful to study society.

Disappointing climax / Where the two political scientists are heading only becomes clear at the end of the book, although the attentive reader will not have missed the accumulating signs. Achen and Bartels want us to normatively embrace the power of groups over the individual and to accept that democracy is a contest among groups. It’s is not just that individuals are inevitably group puppets; it is also good to empower groups, and this provides a new justification for democracy.

Achen and Bartels do not see that the rational voter votes his opinions, not his interests (except to the extent that his interests shape his opinions and whims).

What is needed for democracy, they argue, is to enforce equality between the political groups, although perhaps it is between individuals—they’re not entirely clear. Campaign contributions must be controlled to make sure that some groups, such as the rich, don’t have more political influence than others, such as the poor. More than 300 pages (most of which admittedly are interesting) to come to this normative anticlimax!

But the authors have a problem. If groups are made equal irrespective of their sizes and other factors, individuals will be unequal. If, on the other hand, all individuals are valued equally, their groups—if freedom of association is protected—are unlikely to be equal. As usual, the mirage of material equality (equality of results) leads to absurdities.

Consider another instance. Achen and Bartels talk about “economic and social equality” (emphasis in original)—the whole package it seems. “The most powerful players in the policy game are the educated, the wealthy, and the well‐​connected,” they write. They note that ideally “both corporate interests and college professors would get less weight in the democratic process.” Check your bright privilege, as it were. But if college professors have too much political influence, then Democracy for Realists is a book that should never have been. As an unrepentant sympathizer for old‐​fashioned classical liberalism (“antiquated in its ideas,” as Achen and Bartels would say), I still think that would have been a loss.

The scope of democracy / This drift of Achen and Bartels’ otherwise interesting book suggests that the fundamental problem of democracy as we know it is not ignorant voters, lurking bureaucrats, or people who get a bit more ink or electrons. The fundamental problem is the scope of democracy. If democracy is “the rule of the people,” as Somin defines it in his 2013 book, the first problem is the rule. The second problem is identifying who “the people” are. Moreover, if everything is up for grabs, the volume, depth, and antagonism of discussion will exclude most people. Politics begets conflict and ultimately renders discussion impossible.

“Paradoxically,” writes Somin, “the best way to improve democratic deliberation may be to rely on it less. … Democratic control of government works better when there is less government to control.” Going back to the problem of voters’ rational ignorance, “the problem of political ignorance may be more effectively addressed not by increasing knowledge but by trying to reduce the impact of ignorance.”

Or look at it from Buchanan’s constitutional perspective (as I understand it). We must distinguish between consumer tastes, which are a private matter, and political opinions, which—except in (most) libertarian opinions—imply imposing one’s preferred lifestyle on others. Democracy, even in its direct forms, may be good for issues relating to tastes for public goods. Think of national defense or even public land if you take John Locke’s proviso seriously (private appropriation of land “at least where there is enough, and as good, left in common for others”). But no democracy, bureaucracy, or any other form of “-cracy” (which is derived from the Greek kratos, meaning “power”) is good when tastes concerning nonpublic goods are concerned: Joe’s preference for white chocolate and Alice’s preference for dark chocolate are nobody’s business.

Both Escape from Democracy and Democracy for Realists are interesting books. But the latter is not consistent with real limitation of state power.