An Economic Theory of Democracy

“Although governments are of crucial importance in every economy, economic theory has produced no satisfactory behavior rule for them comparable to the rules it uses to predict the actions of consumers and firms. Our thesis attempts to provide such a rule by positing that democratic governments act rationally to maximize public support. By rational action, we mean action which is efficiently designed to achieve the consciously selected political or economic ends of the actor.”

Summarizing the overall purpose of the book, Downs is proposing a theory that will bring economics and politics in line with practical behavior and observation. Classical economic theory considers politics to be an entirely separate sphere, imagining the free market as an autonomous sphere operating by its own laws, so that political involvement in the economy can only constitute interference. Since governments in the mid-twentieth century had undertaken an expansive role in the economies of their countries, economics itself needs to drop its laissez faire assumptions and see how the logic of politics might fit its own theoretical precepts.

“From the self-interest axion springs our view of what motivates the political actions of party members. We assume that they act solely in order to attain the income, prestige, and power which come from being in office. Thus politicians in our model never seek office as a means of carrying out particular policies; their only goal is to reap the rewards of holding office per se. They treat policies purely as a means to the attainment of their private ends, whjch they can reach only by being elected. Upon this reasoning rests the fundamental hypothesis of our model: parties formulate policies in order to win elections, rather than win elections in order to formulate policies.”

In the Western philosophical tradition, the use of reason is closely tied to the advancement of self-interest. This does not mean that people are selfish or unconcerned with the welfare of others, as someone may just as easily find their interest in charitable giving as in accruing vast wealth and power. However, interest prevails, whatever form it takes, and so the theory holds that anyone seeking political office is fixated on the benefits of that office. This does not prevent them from doing good work for their constituents, but the theory precludes that from serving as a primary reason for running for office.

“When a man votes, he is helping to select the government which will govern him during the coming election period…therefore as we have just shown, he makes his decision by comparing future performances he expects from the competing parties. But if he is rational, he knows that no party will be able to do everything that it says it will do. Hence he cannot merely compare platforms; instead he must estimate in his own mind what the parties would actually do were they in power.”

In the early chapters of the book, Downs makes exceptionally high demands of the voter in order for them to be “rational.” They must know the issues, how positions on each issues would affect them, each party’s position on the issues, and be able to make a realistic assessment of how capably each party can deliver on its own platform. To compound this, the voter must be able to engage in hypothetical reasoning on how the opposition party would have performed had they been in a similar position to the party in power, since platforms alone are not sufficiently instructive. Subsequent chapters will illustrate just how difficult, and unrealistic, such expectations are: at this point Downs is laying out the frame of reference for pure rationality.

“Thus we cannot judge how passionate a majority is by its feeling about any one issue. The members of a passionate majority may only care slightly whether alternative M is chosen rather than alternative N; while the minority may frantically desire N. The crucial point I whether the citizens in the majority have a greater preference for their position than they do for minority positions they hold on other issues. Thus parties do not judge passion by comparing voters with each other; instead they compare the intensity of each voter’s feelings on certain issues with the intensity of his feelings on others.”

Here, Downs notes that passionate voters could conceivably have an advantage because they are willing to work harder than the indifferent voter to communicate their preferences. However, under terms of total rationality, governments are capable of discerning with total precision how much enthusiasm translates into actual votes. A passionate voter means nothing more than a less passionate voter in a system of one vote per person. The voter is only meaningful to the extent that their passion on one issue or another informs their own (or others’) overall decision to vote, and in that respect, voters are all equal regardless of their enthusiasm.

“Thus a man may know the monetary structure of a country without being informed on the present level of the interest rate, the outstanding money supply, etc. Lack of contextual knowledge is ignorance, which is to be distinguished from lack of information. To combat ignorance, a man needs education; whereas to combat lack of information (if he already has knowledge) he needs only information, which is less expensive than education but still costly.”

A democracy frequently touts the value of a well-educated citizenry, and so Downs clarifies that education has a particular value for the voter. Education is limited in its ability to provide specific information on the issues of the day—in general, it can only provide the foundational knowledge from which one can effectively learn about such issues. The accumulation of information depends on time and experience, which one can acquire even without an education in the topic in question.

“Wherever men can be influenced, other men appear whose specialty is influencing them; so it is in our model. Uncertainty renders many voters willing to heed leaders who seem to know the way toward those social goals the voters hold. Thus they follow the leaders’ counsel about which government policies to approve of and which to oppose. Subtler forms of leadership insinuate themselves into the reporting of news, the setting of political fashions, and the shaping of cultural images of good and evil.“

Once Downs introduces the variable of uncertainty, his model undergoes a decisive shift. The interaction of purely rational actors to maximize their interests gives way to a recognition of significant disparities in power and influence. Downs recognizes that government officials have more advantages in swaying the public to their view than the public does in influencing the government. The government not only has an advantage in communicating particular messages to the media, but over time can shape the basic ideas that inform the public conversation.

“Voters in a democracy do not have equal influence on policy formulation even though each has only one vote. Possession of resources other than a ballot definitely increases a citizen’s’ potential influence upon government policy. Active membership in an organization claiming to represent many voters may even further augment this influence. These are not new conclusions; the only novelty is that we have shown them to be the necessary outcome of rational action on the part of the government and its citizens.”

Just as candidates are looking to maximize the perks of the job, so too do voters want officeholders to provide them with the most possible benefits. Like officeholders, citizens want different things—some want low taxes to secure their own wealth while others favor high taxes for projects they favor. But even though governments have considerable freedom when in office, they have to cater to a wide variety of interests. The voter, by contrast, is making a simpler choice between a limited number of parties. The voter therefore has to take into account the limits on a government’s power—a difficult skill for even the most dedicated political observer.

“A citizen may decide for whom to vote by means of ideologies rather than past records. Instead of comparing government behavior with opposition proposals, he compares party ideologies and supports the one most like his own. Thus he votes on ideological competency, not on specific issues. Such behavior is rational in two situations 1) having informed himself reasonably well, the voter cannot distinguish between parties on an issue basis, but can on an ideology basis, or 2) he votes by means of ideologies in order to save himself the cost of becoming informed about specific issues.”

The purely rational voter should be able to understand every single issue on its own merits, as well as reasonably anticipate how each party will perform on all of those particular issues. The introduction of uncertainty reveals how unrealistic such standards are in their minutiae, and so ideology emerges as a shorthand for voters, even well-informed ones, to incorporate a host of policy positions into a consolidated worldview, simplifying the choice between one party or another.

“Because integrity is thus essential to interpersonal relations, rational men come to value it per se. A perfect liar and a perfectly honest man are equally reliable, but almost all ethical systems honor the latter and chastise the former. This valuation occurs in part because communication in a society of honest men is cheaper than in a society of liars. Similarly, in politics men rationally prefer parties which are honest to those which are not, ceteris parabus. As a result, competition tends to force all the parties in our model to be relatively honest.”

This passage is characteristic of Downs’s lack of cynicism and his theory’s balanced approach to self-interest as a democratizing process. He asserts that there is a place for integrity in politics: Voters do want politicians who are generally honest, and politicians still find it in their interest to avoid a reputation for dishonesty, even if they have a periodic interest in skewing the pure and complete truth.

“If our reasoning is correct, voters in multiparty systems are much more likely to be swayed by doctrinal considerations-matters of ideology and policy-than are voters in two-party systems. The latter voters are massed in the moderate range where both ideologies lie; hence they are likely to view personality, or technical competence, or some other nonideological factor as decisive. Because they are not really offered much choice between policies, they may need other factors to discriminate between parties.”

When there are only two choices, ideology is unlikely to be the determining factor because it is too easy for one party to make minor adjustments in its position to exploit what they perceive to be a weakness in the opposition. In a multiparty system, each party needs to find ways to distinguish itself among a variety of choices, and there is also a greater chance of mobilizing a relatively smaller pool of voters around a more stridently partisan position. One of the biggest differences among democratic systems is whether or not a party has to win a minority to win, or whether it simply needs more votes than the others.

“Political rationality leads parties in a two-party system to becloud their policies in a fog of ambiguity. True, their tendency towards obscurity is limited by their desire to attract voters to the polls, since citizens abstain if all parties seem identical or no party makes testable promises. Nevertheless, competition forces both parties to be much less than perfectly clear about what they stand for. Naturally, this makes it more difficult for each citizen to vote rationally; he has a hard time finding out what his ballot supports when cast for either party.”

A two-party system is paradoxical in that the parties have an incentive to resemble one another (assuming that the electorate itself has a relatively even ideological distribution), but they cannot communicate this fact clearly to voters without undermining their own efforts to get elected. Ideological obscurity is an attempt to square the circle, using rhetoric to erect an illusion of stark choice while the election is in fact being decided on the margins, so as to give voters a reason to vote with enthusiasm without pushing positions that will divide them.

“Rational voting in a multiparty system is both more difficult and more important than in a two-party system. It is more difficult because the possible outcomes are more numerous, and it may not be clear to a voter just what his ballot is supporting when he casts it. Yet each voter is more important because the range of alternative policies offered to voters in a multiparty system is likely to be much wider than in a two-party system…in a multiparty system, the victory of a party at the end of the scale opposite to a voter’s preferences may usher in policies he bitterly opposes.”

Downs presents many paradoxes about democratic government: One of the most important is the inverse relationship between the importance of a rational vote and the difficulty of making one. When there are enough parties to increase the value of each individual vote, the task facing even the most enthusiastic and well-informed voter is formidable. Voters are in an even more difficult position because the disparity between the parties heightens the consequences of the election, and given the multitude of options, there is a smaller likelihood of even the most rational voting securing the desired outcome.

“In a democracy, political power is theoretically the same for all men; i.e., each supposedly has the same opportunity to achieve his goals as does every other. Thus the irrationality inevitable in any society-i.e., the inability to achieve goals perfectly-is shared by all men: no one can achieve all his goals. In short, every citizen of a democracy is necessarily somewhat irrational in the pure sense.”

This passage summarizes the most profound irony of the book. Within the framework of rationality, the impossibility of optimal conditions for the exercise of rational functions means that inequalities will emerge which curtail different people’s ability to pursue their interests rationally. To pretend otherwise, that the public sphere really is equitable and that anyone can advance their interests with the right knowledge, is actually behaving irrationally, pushing rationality past its limits in the real world.

“Governments are vitally concerned with the effects their actions have upon the future utility incomes of voters, since voters often decide how to vote on the basis of the prospects for such future income. But a government cannot trade present votes for future votes the way a voter can trade present income for future income. Therefore government has no discount rate of its own to apply to its own income- an income measured in votes.”

Businesses are ultimately concerned with profits, and profits are a readily available asset at any time in the present, the near future, or the distant future. A restaurant can close for repairs, shutting itself to customers in the expectation that revised facilities will eventually compensate for a period of pure loss. Governments and political parties likewise want to persist into the future, and not merely win the next election, but they can never abandon their short-term goals for the sake of longer-term ones. Sitting out the next election to reevaluate strategy or consolidate resources will only harm their chances in future elections, since they have given voters less information with which to evaluate them.

“The slightest degree of uncertainty prevents a democratic government from carrying out all the optimum-furthering actions that are necessary to bring about a Paretian optimum in even a perfectly competitive economy. Because it lacks perfectly knowledge about voters’ abilities and utility functions, a vote-maximizing government inevitably prevents individual marginal equilibria from occurring. Therefore certain arbitrage in the form of certain political bargains could make everyone better off, or some better off and none worse off. But these bargains cannot be made. Free enterprise cannot make them because they involve collective goods or nonmarket interdependencies or both. Hence influencers offering political bribes always find it more profitable to injure some affected citizens than to bribe all of them. Realizing this, a majority of citizens unite to make bribery illegal because uncertainty causes each to fear he may be in the injured minority.”

Bribery and other forms of corruption are generally seen a moral problem, a violation of democratic ideals that the citizenry as a whole will find objectionable. In this passage, Downs explains that, for some citizens, bribery and other forms of influence-peddling are an entirely rational means of communicating one’s preferences to a government that otherwise will have trouble discerning those preferences and weighing their importance for reelection. The bribe itself is less valuable than the commitment it signals from the voter, since the candidate wants the perks of office less than a discrete material benefit. Laws against bribery are thus the way for those lacking influence to protect against those who possess it, and so it establishes a partial safeguard against the inequalities within the electorate.

“A rational voter chooses his selection principles by experimentally sampling the reporting of several different information sources simultaneously. His sampling should cover reporters with widely different selection principles; e.g. a man might read the New York Times, the Daily Worker, and the Chicago Tribune and compare their reports. Next he makes hypothetical decisions on the bases of each reporting sources’ output. Then, as the real situation unfolds, he evaluates the outcomes each of his hypothetical decisions would probably have led to. The selection principles which consistently lead him to make decisions with outcomes closest to his favorite social state are the principles it is most rational for him to use.”

Downs’s counsel is relevant in the present day, where media outlets cover an increasing breadth of viewpoints. Contemporary research upholds his conclusion that voters who consult a wider variety of sources are better-informed than those who consult fewer, even if they spend more time consuming those sources. Downs’s analysis also confirms the current problem of people tending to gravitate toward media that confirm preexisting views (confirmation bias), but whereas this is generally regarded as a problem in need of solving, Downs regards it as an entirely rational outcome of people seeking to acquire information as efficiently as possible.

“Political decision-making in a large-sized democracy cannot be undertaken without fantastic costs unless 1) information is gathered for the many decision-makers by a few specialists and 2) the information each citizen receives is prefocused upon the differential areas of decision. Both of these general conditions must prevail before individuals can begin reducing their personal data costs to match their personal returns from information.”

Among the many paradoxes that Downs introduces in the book particularly Part 3, one of the most significant is that, in order for there to be even a semblance of a well-informed citizenry, there needs to be a pool of experts responsible for providing that information to the masses. There is no way for the overwhelming majority of citizens, even highly motivated ones, to gain information without the existence of experts in journalism, academia, and government whose main function is to provide it. Public information therefore depends on an oligarchy of experts.

“Since the mass media of communications in many democracies are owned or dominated more by high-income interests than low-income ones, low-income citizens are more likely to receive data selected by principles conflicting with their own than are upper-income groups. We do not know how great an effect this conflict has. However, it contributes to the general advantage of high-income groups produced by the necessity of bearing costs to obtain political information.”

Downs’s portrayal of informed voting is further problematized here when he considers that the experts chiefly responsible for providing information to the electorate are themselves likely to reflect the interests of a small minority of citizens whose preferences are quite distinct from those of the average news consumer. The most likely result, he argues, is that the average reader will either realize such sources do not reflect their interests—in which case they will no longer consult them—or they will mistakenly believe they are being informed on how to pursue their own interests when they are in fact pursuing the interests of those who already have every advantage in bending government to their will.

“A rational man may buy political information because 1) he wishes to influence the government’s policies, 2) his prediction of how other voters will act indicates that the probability is relatively high that his own vote will be decisive, or 3) he derives entertainment value or social prestige from such data. Nevertheless, in so far as voting is concerned, we believe that it is rational for a great part of the electorate to minimize investment in political data. For them, rational behavior implies both a refusal to expend resources on political information per se and a definite limitation of the amount of free political information absorbed.”

Rationality enjoins the pursuit of self-interest, but as the text progresses, self-interest comes to encompass more than the maximization of material or even ideological benefits. This includes a cost-benefit analysis. A person’s time is precious and spending too much of it in deciding to vote is not in their interest. A commitment to democracy may still result in a vote, but in such cases the vote is speaks to a need to enact democratic process rather than an effort to secure a particular outcome. This remains rational.

“Democracy is often defined as ‘government by consent of the governed.’ We can further define ‘government by consent of the governed’ as ‘decision-making in which the decider makes each choice on the basis of preferences of each in proportion to the degree to which he is affected.”

Downs’s refusal to embrace a normative theory his highly analytical and demonstrates his adherence to the “positive” theory model. In order to make his theory reflect the real world as much as possible, he needs to draw a precise distinction between how democracy understands itself and how it works in practice. The idea of self-government is not a sham, as governments and voters each retain the power to decide for themselves but, ultimately, rational action must have a meaningful connection to self-interest. For the overwhelming majority of voters, a decision will not affect them in a significant way: Rationality predicts the minimum proportionate contribution.

“The amount of information a man has necessarily affects the confidence with which he holds his decisions, but it does not necessarily affect their nature. If everyone had 100 percent information, some citizens might still be indifferent. Therefore indifference is not merely an illusion caused by lack of data; so we cannot argue a priori that increases in data will tend to eliminate it. However, more information does raise the confidence of each citizen in his decision, ceteris paribus, because it moves him closer to being 100 percent informed. For this reason, the more data a man has, the less he must discount his estimated return from voting correctly.”

Conventional wisdom holds that high levels of voter indifference stem from a lack of information and, while there are some correlations between levels of education and political engagement, Downs does not equate either education or access to specific information with political engagement. His theory posits a deeper level of inherent political equality. The most devoted consumer of politics may have learned enough to conclude that politics itself is a corrupt affair with no meaningful differences among the parties, or may simply enjoy learning and talking about politics with no concern for the outcome.

“Economic theories of government behavior-in so far as they exist-universally fail to assign any motives to the men in government…the theorist discussing government’s role in society merely describes its proper function. He says nothing about the incentives which might cause that function to be carried out by the men who run the government. Yet those incentives are vital, because their operation determines in what way the function of government is discharged, just as the degree of competition among firms determines what goods are produced.”

Economists have created an overly artificial separation between their field and that of politics, based on the assumption that theirs is a field that emerges spontaneously from social interaction whereas government is an artificial creation with its own idiosyncratic set of rules. Downs revises this with the common-sense conclusion that the economic and political spheres are similarly populated by human beings who do not fundamentally alter their nature when they operate in one environment or another. The incentives change based on circumstances, but there is a baseline human motivation which supersedes the particular functions of either field.

“Since constitutions vary widely, this rule is not the same in every society. The behavior of government in a democracy containing many competing parties is bound to differ from the behavior of a government in a one-party totalitarian state. Nor can this disparity be expressed merely as a different weighting of some general welfare function. In this case, the very processes of social action are so unlike that any theory which tries to encompass both of them must either self-contradictory or too general to be meaningful.”

In accordance with the behavioralist tradition, Downs keeps his model as general as possible, so that its hypotheses are not contingent upon the particularities of any regime type or institutional design. Even so, structures matter, and different kinds of governments will structure incentives in different ways, sometimes radically so. As desirable as it is to have a single mode of behavior informing all of politics, Downs counsels his fellow economists to be mindful of significant differences within political life, lest their models achieve consistency at the cost of accuracy.