Administrative Democracy?

Guest Blogger

For the Balkinization symposium on Susan Rose-Ackerman, Democracy and Executive Power: Policymaking Accountability in the US, the UK, Germany, and France (Yale University Press, 2021).

John Ferejohn

                Democracy and Executive Power is an important study of the administrative state in advanced democracies.  It usefully contrasts practices in France, Germany and UK – all of which have long and deep experiences with public adminstration --  with those in the United States.  I think the most important contribution is her assessment of the administrative state as an essential locus of democratic deliberation.  She thinks administrative procedures must aim to integrate the exercise of delegated authority with expertise and open ended consultation with varied social interests.  While the other administrative states in her study try to reconcile the exercise of delegated authority with technical expertise, the American regulatory state is more ‘democratic’ or open to external social interests.  Rose Ackerman sees many flaws in the American approach but seems to regard the openness to external voices as something that needs to be expanded, improved, and exported to other democracies.  I am very sympathetic with this project and have defended (elsewhere) the deliberative democratic role of administrative agencies in American democracy.

                I read the book, therefore, as seeking to “democratize” administrative agencies in order to increase their capacity to make rules that are both (technically) good and legitimate (publicly acceptable). I understand the first criterion (good rules) as being in Rose-Ackerman’s terms outcome oriented whereas the legitimacy requirement is presented as a procedural criterion.  I am not convinced that it is possible to make such a clean separation between outcomes and process.  In the book, she see outcome based evaluations as external to the agency: the courts insist that agency rules comply with the the constitution for example.  Presidents have insisted repeatedly that agency rule making process should be checked by cost benefit analysis. And Congress (backed by courts) requires that the authority for agency action be found in congressional statutes (which impose both outcome and process based requirements). Of course agencies make their own internal outcome based evaluations: the FDA evaluates new drugs in terms of safety and effectiveness for example. Still, she wants to argue that outcome based evaluations are insufficient to justify agency rule making and that the use of good procedures form an independent basis of evaluation.

Rose-Ackerman rejects the idea that agencies can be restricted from making rules with the force of law by some resurrection of the delegation doctrine. Such a rehabilitation would impose arbitrary limits on agency powers and would effectively limit the capacity of Congress to make policies it decides are needed. She also rejects the idea that the legitimation of agency actions proceeds only transitively from elections through delegation and political appointment. Such a view seems both over and underinclusive in many ways. Rather she wants to see agencies as having somewhat autonomous legitimacy claims to making legislative-like rules.  Such claims, she thinks, must be grounded either on the quality of outcomes or policies, or on the procedures used by the agencies to make and revise policies.

Rose-Ackerman opts (mostly) for procedural rather than outcome based legitimacy and wants agencies to extend and improve their policy making processes – including APA requirements for notice and comment rulemaking – by opening agency procedures to outside input in ways that the agency itself cannot limit. She recognizes, as she must, that open procedures may well lead to distorted representation of public views – dominated by industry or other interest groups. Such biases can, she thinks, be offset by policies encouraging or subsidizing the participation of outsiders (consumers, environmentalists, etc.). She seems to think that Federal agencies in the United States are mostly on the right track, although there is much room for improvement.  But she recognizes that the other countries in her study – France, Germany and the UK – have not gone down this path.  She argues that they should.

The examples of openness that she presents from Europe seems mostly (in my reading) to concern siting of public projects or other distributional issues.  Where to put a pipeline, a nuclear power plant, a overhead wires; things like that. Classic NIMBY problems in which the program itself is not really open to reconsideration based on public input; the only question is who or which communities it will impact. A classic failure to consider political reactions to distributional impacts might be the yellow vest movement in France.  It seems a short step for European administrators to be convinced that they should find ways – procedural ways of the kind Rose-Ackerman favors – to pay attention to things like this.  It seems a bigger step to get European administrator to open their decision processes more widely to consider which public projects should be adopted in the first place.  All of these states have deep administrative traditions – Whitehall administrators in the UK, French Enarchs and Polytechniciens, and the descendants of the Prussian state – all of which present themselves as legitimated indirectly via the electoral process (they are merely enforcing parliamentary statutes; the model Rose-Ackerman rejects) and as bastions of technical expertise and administrative wisdom.  Compared to the American system, the European administrative states have only thin layers of political appointees and are dominated by specially trained administrative elites.  My point is not that democratizing reform is impossible in those traditions. Rather it is that, if such reforms are to take place, reformers need to persuade the administrative as well as political elites. If such things are imposed from outside, reforms may not stick or remain superficial.

Perhaps a caution can be drawn from Macron’s reaction to the yellow vest movement.  Rather than seeking to reform the policy making process, his effort seemed to directed to open a new channel to the public, outside of administrative processes, directly to the president of the republic. This channel took the form of popular assemblies or fora that encouraged citizens to vent their complaints to him.  Maybe this will work for some things but Macron’s effort seems to me to essentially leave the administrative state structures themselves completely untouched and instead to erect on top of it a centralized process for hearing what used to be called doleances.

In the US the problem is quite different.  Here the administrative state is heavily penetrated politically, with armies of political appointees arriving after each election – often long afterwards if the new president’s party does not control the Senate – most of whom stay for relatively short time periods.  Political appointees have the capacity to structure APA processes to favor interests they (or the president) prefers and often sideline disfavored proposed rules in various ways.  This is a politically captured process – to varying degrees of course – and one would expect presidential appointees to resist efforts to open up their policy making procedures to disfavored interest groups. Those appointees who favor an agency’s programs will see such efforts as having the effect (and probably the aim) of lengthening the administrative process and introducing veto rights.  The result is often noted: it is nearly impossible for an incoming president to get his favored agency rules in place before the following presidential election.  If a president wants to make a permanent mark on policy, he or she needs to serve two terms.

I think that the prospects for procedural reforms of the kind Rose-Ackerman favors are pretty limited both here in the United States and in Europe too. I do not deny that agencies have essential roles in making rules with the force of law and that procedures may have the effect of lending democratic legitimacy to the rule making process.  But achieving these gains seems difficult in the world of bad faith that we seem to live in at the moment.  My guess is that agency legitimation is more likely to be produced based on outcome based considerations.  Good policies, even if unpopular for a time, can gain acceptance through subsequent processes of public acceptance.  Such ex post legitimation seemed to be the bet that Obama made with respect to the ACA regime.  But I think it is pretty common.  One can argue that ex post legitimation – through popular use and acceptance – has some procedural aspects. One of the things that people do after policy regimes are established is to make use of new procedures (and old procedures under law).  But they also decide whether they like the outcome itself.  That seems to be the story of Social Security and Medicare and Medicaid or at least part of it.

John Ferejohn is the Samuel Tilden Professor of Law at New York University and the Caroline S. G. Munro Professor of Political Science and Senior Fellow at the Hoover Institution, Emeritus. You can reach him by e-mail at ferejohn@stanford.edu.