Moe: Interests, institutions, and positive theory

Moe. 1987. Interests, institutions, and positive theory: The politics of the NLRB. APD.

Overview

Two independently-developed theoretical approaches--economic (pluralism) and positive theory of institutions--have attempted to explain regulatory outcomes. Both have deficiencies, though both move us considerably in the right direction. However, we must avoid the temptation to find a single body (e.g. Congress) that controls regulatory outcomes. By looking at the NLRB, Moe shows that multiple principles influence regulatory outcomes.

Part I is empirical--a history of the National Labor Relations Board. Part II discusses the lit's two theories in light of the NLRB experience.

The NLRB: A Brief History

Scarce Political Resources and Legislative Stalemate

Political actors have many goals but scarce resources. Business groups would love to have an anti-labor government, but they are also interested in tariffs, taxes, environmental regulation, and so on. Similarly, labor groups would love to have a pro-union government, but they also are interested in other things. Congress and the President also have ideas about what labor policy should look like, but they will not spend time on labor policy unless they foresee political consequences for failing to do so. Thus, the most common state of policy conflict is peace--or cold war--but not active conflict. The NLRB's history reflects this observation.

Origins of the NLRB

The NLRB was born of the pro-labor Wagner Act (1935), then amended by the pro-business Taft-Hartley Act (1947). The period between these two laws was one of active political conflict between business and labor. However, business and labor got along quite well--despite continual efforts to advance their interests--for the next thirty years. Why? After Taft-Hartley, business and labor faced legislative stalemate. Neither side had the political power to push a major legislative initiative. Thus, they consigned themselves to working within the NLRB. They busied themselves mostly with influencing presidential appointment of NLRB board members (one appointment each year to a five-year term).

The Early Years: Appointments to the NLRB Board

The president, meanwhile, did almost nothing to influence NLRB processes. He made an annual appointment, but then let the Board act independently. Similarly, the Senate had bigger fish to fry, so it would have blocked only an obviously unqualified candidate--which never happened (until Reagan--see below). When it was time for an appointment, (if the president was Republican) business would submit its list of ideal candidates. (If the president was Democrat, substitute "labor" for "business" and vice versa.) This list was vetted. The candidates were moderates, albeit moderates who leaned toward business. Why moderates? Because business knew it lacked the resources for a major legislative fight, so it resigned itself to trying to get along with labor. Labor, meanwhile, would vet the list, and object only if business recommended an unusually extreme candidate; after all, labor didn't want to cry "wolf" too many times. Having received the list, the president's people chose the one that seemed to fit the White House's political goals the best. This pattern became the normal way of making appointments.

The Early Years: NLRB Moderation and Professionalization

Meanwhile, the NLRB became quite moderate and professional, for two reasons. First, both business and labor interests were interested in nominating moderate board members (see above). Second, the NLRB quickly became the place that young lawyers wishing to specialize in law would work for a few years to establish a reputation. Professional, legalistic work became a stepping stone to a good law career. Thus, the NLRB became professional, efficient, and generally equitable.

Threats to the Equilibrium: Mid-1970s and on

There are two sources that can threaten this equilibrium. Both occurred in rapid succession.

1. Unemployment, inflation, etc. led unions to place greater value on labor law reform. Thus, perceiving that legislative stalemate might be at an end, labor abandoned "business as usual" and fought aggregisely for some changes. This had a chain reaction, leading business interests to also reject moderate compromises that, in an earlier time, would have passed without any difficulty.
2. Reagan and the new conservatives had an ideological mission that was incompatible with the "hands off" approach of earlier presidents. Reagan's goals required active involvement in every agency, since his mission was to shrink the government dramatically. Thus, he completely ignored business's suggestions when making nominations to the NLRB board, and instead nominated members who were far more anti-union than anybody that business had ever suggested. There are indications that business was quite uncomfortable with this. The threat to NLRB's equilibrium was not the new board members--it was a new variety of politician, one that had strong ideological goals.

Theoretical Discussion

Moe uses the NLRB story to evalute two competing theories, and argue that they need to be combined and reformulated.

Economic Theory (ET)

The economic theory is based on the Stigler-Peltzman model, and is heavily premised on Coase's theory and Olson's model of collective action. Political institutions are a black box. Economists want to know how resources are related with outputs--and they don't look at the "black box" of institutions in the middle of this process (until recently). If Coase's theorem holds (and there are no transaction costs), then it doesn't matter what goes on in the black box.

• Economic interests are group-based. Politics is a struggle among economic groups.
• Resource mobilization: Following Olson's model, there is a mobilizational bias in favor of small groups. Concentrated business interests get what they want more easily than diffuse consumer groups.
• Control of politicians by economic interests: Electoral self-interest leads politicians to represent group interests faithfully.
• Control of bureaucrats by politicians: Assumed to be complete.
• Modeling strategy: Comparative statics. Equilibria are based on governmental allocations of benefits and costs.

Positive Theory of Institutions (PTI)

PTI developed from problems of social choice, to explain how political outcomes can be so stable. Because the approach begin with studies of Congress (agenda setting, legislative rules, etc), it has remained quite Congress-centric. Since transaction costs are rampant, institutions matter.

• Economic interests are individual: Voters have interests, and these interests influence politicians.
• Resource mobilization: Not discussed much. Olson's theory plays much less role. Individual-level interests do not need to be mobilized to influence legislators, apparently.
• Control of politicians by economic interests: As in economic theory, politicians are electorally motivated. Unlike economic theory, this does not imply perfect control of politicians. It requires looking at institutional rules to know how well politicians will represent various groups.
• Control of bureaucrats by politicians: PTI emphasizes problems of Congressional control. PTI's premises, however, seem to point in the opposite direction: Congress should tolerate a wide range of agency discretion, since it takes quite a bit of malfeasance before MCs have electoral repercussions.

=Why the Theories Need Reformulation

1. ET's logical flaw: ET uses Coase's theory (zero transaction costs) to justify not looking at institutions, but it uses Olson's theory (high transaction costs) to explain why politicians listen to group interests. This is a significant contradiction.
2. Similarly, ET uses Olson to explain why groups mobilize, but Olson's theory does not imply that group leaders will faithfully serve group members--after all, they use "selective incentives" (unrelated benefits) to get people to join the organization, so there is no reason to expect group members to punish group leaders for unresponsiveness to group interests (as long as the selective incentives remain in place).
3. Both ET and PTI put too much value in reelection incentives. Clearly, politicians can also have policy preferences. Sure, a House member might need to attend to reelection concerns, but a president--Constitutionally limited to two terms--does not. Reagan's ideology clearly led him to act in ways that were different from all previous presidents--his own preferences clearly mattered.
4. "Institutions" can mean more than formal rules. For example, norms mattered during the peaceful equilibrium of NLRB politics. (Norms of reciprocity, moderation, and so on.) Also, Senate norms ("folkways," perhaps) led Senators to allow the president to nominate just about anybody that wasn't exceptionally extreme, even though Senators surely had electoral incentives to do otherwise.
5. When it comes to political control of the bureaucracy, the principal-agent (PTI) framework is misleading. It leads us to expect an agent (usually Congress) to actually pay attention to control. But the president exercised clear control throughout NLRB history, and all he ever did was make an annual nomination. "Presidents 'controlled' the agency without trying or caring" (p290). Control can be "indirect, unintentional, and systemic" (p291, also note 69 about McCubbins).
6. The agency influence its environment. The NLRB's early years caused major unionization in most large firms. The firms eventually stopped resisting unions and changed to accept them as a way of negotiating with labor. Thus, the NLRB changed business, which therefore changed what it asked of the NLRB.

In Sum

PTI has a much stronger logical basis; ET is seriously undermined by a couple major logical contradictions. Nevertheless, PTI could also use improvement. We need a more systemic theory--one that looks at how actors/institutions influence on another reciprocally, how they change over time, how their institutional relations are both formal and informal (i.e. norms), and so on. PTI leads us to overlook these factors.