Squawking & Capture: Conflicting Theories of Bureaucracy?
They say the road to hell is paved with good intentions—and no matter how well-intentioned, bureaucracies often end in disaster.
One of the core contentions of economists such as Milton Friedman, George Stigler, William Niskanen, James Buchanan, and Gordon Tullock is that policymaker and civil servant sincerity is not a policy outcome determinant. What matters are the institutions they inhabit and the incentive structures under which they operate.
Over the past few decades, economists have also pushed for more formally rigorous theories and empirical strategies capable of identifying causal effects of what the social science field has dubbed the “causal revolution”. Those revolutionary strategies have begun to be applied systematically to bureaucracies only recently.
One useful example of this newer, rigorous approach is Clare Leaver’s “Bureaucratic Minimal Squawk Behavior: Theory and Evidence from Regulatory Agencies.”1 The paper’s major contribution is what the author calls the “minimal squawk” hypothesis: career-concerned bureaucrats may distort their decisions to avoid visible public criticism. Leaver contrasts this mechanism with more familiar stories in the public choice tradition, that is, Niskanen’s view of budget-maximizing bureaucrats and Stigler’s theory of regulatory capture by special interests.2 On these alternative views, the incentives of bureaucrats put them knowingly at odds with public welfare. After summarizing her model and evidence, I argue that these hypotheses need not be mutually exclusive; rather, Leaver’s hypothesis is a refinement that helps us understand how reputational concerns, budget ambitions, and interest-group pressures may interact and reinforce distortions.
Leaver models a three-player game to isolate what is going on between a regulator (the bureaucrat) and a regulated firm. The third player is an evaluator, who might be a manager, an agency head, a politician, a future employer, or even the press. The regulator has policy preferences and gets a kind of “warm glow” from making the right call, but also cares about how smart or competent they look in the eyes of this evaluator. Career concerns influence the regulator to weigh that reputational payoff more or less heavily. The evaluator, for their part, forms a view about whether the frontline regulator is high- or low-ability. The regulated firm wants to keep profits high and costs down, and the regulator’s decisions—through rates, penalties, or other conditions—can raise those costs in ways the firm would like to avoid.
Gameplay runs roughly as follows. The regulator faces a case and must choose between a “tough” or a “generous” decision by using an imperfect test to determine which option is more appropriate. Think of a manufacturing plant subject to limits on how much it can pollute a nearby stream. The regulator samples water quality upstream and downstream to infer whether the plant is violating its permit. Based on that test, they decide whether to impose a strict sanction (tough) or to be more lenient (generous), knowing there is some chance they will either misjudge the situation or impose costs on the public that are not worth the benefits. The regulator has two concerns in mind: making the right call and impressing the evaluator who will later judge their competence. The firm, for its part, can “squawk” by challenging a decision that hurts it (for example, raising a complaint, mobilizing lawyers, or going to the press) and thereby threatening the regulator’s reputation.
The minimal squawk logic falls straight out of this game. Once a regulator cares about her career reputation and the firm knows it, the firm has every reason to commit to making a lot of noise whenever it gets hit with a tough, unfavorable decision and to stay quiet when treated generously. That asymmetric threat changes the regulator’s calculus. Lower-ability regulators, who know their information is less precise, see that tough decisions expose them to a higher risk of being labeled “incompetent” if the firm squawks, so they lean toward generosity and, when reputational concerns are strong enough, sometimes almost always play it safe. Higher-ability regulators are more willing to be tough because good tough decisions are the best way to signal competence. But even they become more cautious as the agency or political environment puts a higher premium on reputation. The upshot is a systematic bias toward generous decisions, and that generosity bias gets stronger as career concerns matter more.
Before turning to the empirical test of the minimal-squawk story, Leaver retools the model to build a clean version of the classic capture hypothesis.3 She makes three changes. First, the regulator now benefits from a transfer offered by the firm—a “bribe” in the formal language of the model. Second, reputational concerns are switched off: decision-making ability is treated as known and fixed, while the only private information the regulator holds is about her intrinsic motivation to “do the right thing.” Third, at the start of the game, the firm commits to pay a non-negative bribe if it receives a generous decision in the low-cost state. Let me stress that a “bribe” here need not mean a suitcase of cash; it can be read more broadly as any transfer, perk, or future favor that makes generosity privately attractive. Compared with the first version of the game, the firm has one of two distinct levers over the regulator: the threat to squawk or the promise of a side payment.
The empirical test pits these two levers against one another. Leaver assembles a matched panel of U.S. state public utility commissions (PUCs) and investor-owned electric utilities, and then estimates how a commissioner’s statutory term length affects the incidence of formal rate reviews. In her setup, the minimal squawk story predicts that longer terms should increase the probability that a firm faces a review, especially in periods when operating costs are falling. The idea is that short terms make commissioners more reputation-sensitive and therefore more inclined to avoid tough, noisy reviews. The capture variant predicts the opposite: if longer terms mainly facilitate cozy relationships with regulated firms, then longer terms should decrease the probability of review, since more entrenched commissioners find it easier to trade generosity today against favors tomorrow.
Leaver’s first cut at the data asks which lever seems more consistent with how regulators behave. She looks at states that shortened PUC term lengths over the 1980s and compares them to states that kept long terms. In the “switcher” states, the share of utilities facing a formal rate review fell from about 64 percent in 1982 to roughly five percent in 1990, while in “constant-long” states it fell from about 57 percent to 25 percent. The resulting difference-in-differences estimate (around -26% points) is statistically significant at the 10 percent level, and an event-study around term-length cuts shows a clear downward shift in review incidence after terms are shortened. In other words, when PUC terms get shorter, reviews become less frequent.
In her concluding remarks, Leaver writes that it is “fair” to reject the capture hypothesis in this setting, but that it is a “thornier question” whether the results uncover minimal squawk.4 One obvious alternative is learning: commissioners with shorter terms may review less often simply because new appointees need time to learn the ropes, or because experienced commissions face a lower marginal cost of launching a review. Regardless, I do not think these findings overturn the broader capture hypothesis in public choice; at most, they show that, for this slice of U.S. utility regulation, reputational pressures provide a better organizing lens than straightforward quid-pro-quo bribery. In what follows, I’ll explain why.
I do not want to spend too long on the econometrics here, but it is worth noting that Leaver’s empirical test rests on several strong design assumptions. She assumes, for instance, that a firm’s propensity to file for a rate change is conditionally independent of statutory PUC term length once you control for observable firm characteristics. It is not obvious why we should expect that to hold: firms may well adjust their “squawk” strategies in anticipation of how exposed short- or long-term commissioners are. She also treats statutory term length as a clean proxy for both the strength of reputational concerns and the shadow cost of transfers, and as at least sequentially exogenous, given a rich set of institutional and political controls. Again, it is not obvious why reputational concerns and the cost of providing transfers should move in neatly separable directions. Shorter terms also narrow the window in which a commissioner may receive favors, which could make those favors more valuable at the margin, and they might reduce concern for reputation with some evaluators while increasing concern for others (what if the relevant audience is the regulated firm itself?). Finally, the empirical specification relies on a specific proxy for the “low-cost” state—lagged declines in operating expenses—that is only valid under fairly restrictive assumptions about investment dynamics and the fixed costs of initiating a rate review.
Taking the assumptions and the headline results at face value, the capture variant in the model corresponds to a very specific channel: firms offering transfers in exchange for generosity with no role for reputation.
Once we frame it that way, however, it becomes less clear that the data really discriminate between “minimal squawk” and what public choice theorists usually mean by capture. In practice, capture is not just suitcases of cash; it is differential influence. Larger or more politically connected firms have more at stake, more capacity to litigate or lobby, and more ways to reward cooperative regulators. A test of that richer capture hypothesis would therefore ask whether changes in term length disproportionately affect review probabilities for large versus small utilities, or for firms with more residential customers or higher revenues. Leaver’s design does not exploit this kind of heterogeneity. And even if we did find, say, that commissioners with short terms review many small firms while avoiding noisy fights with the largest utilities, it is not obvious whether we should label that as “minimal squawk” or “capture,” or both: reputational incentives and interest-group leverage are working through the same channel of asymmetric pressure on regulators.
This is why minimal squawk ultimately complements, rather than replaces, capture. Larger firms pose the greatest squawk risk to regulators simply because they have more resources and more channels through which to complain, litigate, and lobby. Minimal squawk gives us a language for that asymmetry in reputational pressure. But it should not start from the premise—suggested by some survey evidence—that bureaucrats are otherwise straightforwardly public-spirited, while the capture hypothesis simply “gets it wrong.” In many settings, the regulated firms themselves are the most important evaluators: they can shape commissioners’ post-agency career prospects, reinforce or undermine their reputation in professional networks, and sometimes directly hire them after their appointments end. A richer theory would trace how these reputational and material channels interact, treat “squawk” and “capture” as two sides of the same institutional coin, and then rigorously test the joint implications rather than pitting them against each other in isolation.
[1] Leaver, Clare. “Bureaucratic Minimal Squawk Behavior: Theory and Evidence from Regulatory Agencies.” American Economic Review 99, no. 3 (2009): 572-607. http://www.aeaweb.org/articles.php?doi=10.1257/aer.99.3.572.
[2] Niskanen, William. Bureaucracy and Represntative Government. Chicago and New York: Aldine & Atherton, 1971; Stigler, George. “The Theory of Economic Regulation.” The Bell Journal of Economics and Management Sciences 2, no. 1 (1971): 3-21. https://doi.org/10.2307/3003160
[3] For a formalization of the capture hypothesis, see Laffont, Jean-Jacques and Jean Tirole. “The Politics of Government Decision-Making: A Theory of Regulatory Capture.” Quarterly Journal of Economics 106, no. 4 (1991): 1089-1127. https://doi.org/10.2307/2937958
[4] Leaver, “Bureaucratic,” pg. 599.
Stay Informed
Sign up to receive updates about our fight for policies at the state level that restore liberty through transparency and accountability in American governance.