Thank you, Chris, for that very warm introduction and thank you to AEI for having me here today and to AMGEN for sponsoring this forum.  I would also like to thank the CSPAN viewers from across the country for tuning in today.

It is humbling to come to AEI to give an AMGEN forum on regulation.  This institution throughout its history has stood for the highest standards of academic research on regulation.  Practically all the scholars here at AEI work on regulation in some capacity or another. I would just like to mention a few whose research overlaps perhaps most closely with my kind of simple minded work today.  Chris DeMuth your work on the expansion of government regulation and your recent addresses on this topic; Bob Hahn on national regulatory structures; Greg Sidak and Tom Hazlett on the subtleties--and not so subtleties--of telecommunications regulation.

Government officials often think that academics don't really understand what the real world is about, and academics often understand that government officials don't really understand what the real world is about. AEI provides a very  good bridge for the two.  Having a foot in both worlds, I probably don't understand much of anything. 

I have thought about writing a book on regulation.   Someone recently put me in touch with a leading book agent on regulation. This was the person who was the most sympathetic to the idea of a book on regulation. I called him up for a lunch appointment to discuss this book possibility. He said, "First of all, tell me about this book you want to write." I said, "Well, it's about regulation."

He said, "Thank you, but really there isn't a market for books on regulation. No one wants to read about regulation."

I tried to deepen my voice and sound important. I said:

"I really have some new ideas." He said: "Save yourself some trouble, kid. Photocopy your pages, and hand them out to your ten closest friends. No one else wants to read this book. If you have a book on sex or health, something like that - - that's a book I can sell. But regulation -- never!"

I'm just glad I didn't talk to one of the agents who was not sympathetic to the idea of a book on regulation.

People in government sometimes have an inflated sense of self-importance. Much of the outside world really doesn't care about what goes on in government. I once met with a venture capitalist in California. He asked me what I did. I said that I was an FCC Commissioner.

He said, "That’s interesting. I have a friend who is on some blue ribbon panel in Washington. He flies there once a quarter and has a good time. So tell me, what's your real job?"

I said, "Well, I am an FCC Commissioner." The venture capitalist was startled to discover this was a full time job. A lot of people in Washington are startled too. While the outside world doesn't know much about what regulators do, those who have to deal first-hand with regulators know all too well.   

There is a story of the regulatee who had been called to a meeting with a regulator. You can pick your favorite regulator to fill in the blank. This regulatee was in a lot of trouble, so the company sent a junior officer to go to the meeting.

The regulator just grilled this corporate officer all day long: "So, have you filled out these forms and those forms? Do you know how to fill out the forms?" The officer did not know the answers to any of the questions. And at the end of the day, the regulator turned to the regulatee and asked: "What are you going to do?" 

The poor officer said, trembling: "Well, I've no idea what you've told me to do. I don't understand what I'm supposed to do, but I'm sure the good Lord will see me through this."

The regulator shakes his head and goes in to see one of his colleagues.  He says, "Well, I have some good news and some bad news. The bad news is: this corporation has no idea what they are supposed to be doing. They really don't understand. The good news is: he recognizes that we are actually God."

I am going to look at regulation as an economist would.  But at times there is this element of god-like nature of the regulators that is troubling. That is the theme of my talk today--as Chris so eloquently pronounced--the realpolitik of regulation.

I am also going to suggest that there is a market for regulation with demand and supply characteristics. But I am also going to suggest that pure economic analysis alone can only take us so far in understanding regulation. We also need to understand some institutional and political considerations of the regulators themselves.

One of the main themes of the talk is that regulation today is monopoly provided. There is a government-sponsored set of regulators, and we all have to deal with them. We cannot choose a different set. The monopoly position of regulators gives them substantial power in regulated markets, all of which may harm both competition and consumers in the market.

Regulators expand their power in markets by enhancing their discretion to make market-affecting decisions through a series of techniques, the realpolitik of regulation. I am going to list ten of these which will be mentioned throughout the talk. These are:

a) Lack of strict adherence to the law

b) Expansion of agency jurisdiction

c) Refusal to write procedural rules

d) Refusal to have transparent processes

e) Refusal to have predictable processes

f) Refusal to collect, disseminate, analyze, or systematically evaluate information on the costs and benefits of regulation

g) Complexity

h) Inefficiency

i) Promotion of benefits and hiding of harm to consumers

and (j) Treatment of similarly-situated parties in different ways.

Each of these ten attributes contributes to the ability of a regulatory agency to make discretionary decisions that may affect regulated markets. The offensive strategy for regulators is the realpolitik of regulation is to expand these characteristics;  and to use them as much as possible. The defensive strategy for the rest of us is to diminish these powers. 

Let me give a disclaimer--

DISCLAIMER: I will speak about regulation generally. By regulation I mean government decisions that restrict what can be provided in a market. It could occur at any level of government, by any agency, and by any level of employee of the agency. Many of my examples will come from the FCC because that is the agency that I know best.  But it is by no means the only one that these principles apply to.. My comments are academic in nature in nature, as befits an academic institution that's hosting this event. They are not intended as a criticism or endorsement of any particular agency, or any particular person.

The employees of government agencies in general, and the FCC in particular, have, to the best of my knowledge, the highest integrity and standards. Chairman Kennard of the FCC has, in many areas, done much to limit agency discretion, and I commend him for his efforts. But all of us, as regulators and employees of regulatory agencies, must understand how best to administer the agencies the public has entrusted with us. In that spirit, I offer these comments which apply equally well or poorly to the FCC and to every other government agency.

Before I describe in more detail these 10 characteristics of the realpolitik of regulation, let's examine briefly the economics of the market for regulation. What does the demand for regulation look like?

Demand:

There are at least four sources of demand for regulation: consumers, businesses, support industries, and bureaucrats. Consumer demand for regulation is the public finance, textbook description of regulation. That is, regulation is mechanism to cure market failures and to internalize externalities. Thus, for example, we have environmental regulation to provide for a cleaner environment, because there are externalities to environmental quality, and many consumers if not all consumers want a cleaner environment.

Or there may be non-market failures such as high transaction costs, high enforcement costs, and high litigation costs. Thus, we have a county recorder of deeds not because there is an externality or market failure that prohibits a purely private market solution, but because the transaction costs of deed recordation, and enforcement costs associated with property and contracts, may be lower with governmental recordation of deeds.

Similarly there may be cases of peculiar distributions of risk in a market, and the government steps in with risk-sharing mechanisms, such as government-subsidized health care. These may or may not be better market-based solutions, but my discussion today is not premised on whether or not a specific regulation meets some form of economic rationality. Indeed, the general result is that all of the above forms of government intervention can be rationalized in a world in which common law, property rights, and contract rights are paramount, and in which government intervention is merely there to enhance some form of quantifiable social welfare where the benefits of regulation demonstrably exceed the costs. Economists will quibble about how best to quantify these benefits and costs, but today I simply want to discuss something different.

The second source of the demand for regulation are the bureaucrats themselves.  Folks like me. Folks like employees of the federal government-- a lot of jobs at stake.  Also, a lot of power. That's part of the theme of the realpolitik of regulation.

The third source of demand is the service industry of lawyers and accountants that demand regulation their livelihood depends on regulation. The more regulation and the more complex the regulation, the more the demand for the service industry to interpret it. 

Finally, businesses also demand regulation, they want regulation favorable to themselves and unfavorable to their competitors. In Washington, it is called the "level playing field."  Much academic work has been directed at all of these sources that have a demand for regulation, and I am not going to dwell on them today.  What I am going to describe briefly are what an economist would describe as the factors affecting the demand for regulation.

The demand for regulation depends on various price effects both its own price and price of alternatives certain quality characteristics and income.  I am not going to belabor all of these today. They are the topics of chapters of the book that won't be written because no one will read it.  I will, however, discuss in some detail the characteristics of  the demand for regulation.

Let me just note in passing that the own price or cost of regulation is substantially different for different individuals in an economy. For some consumers, who do not drive automobiles, a regulation such as a CAFE fuel efficiency standards is a regulation that may be costless. For others, who drive for many hours each day, it may be extremely costly. Consumers like those regulations that favor them individually, and hate those that harm them individually, and are indifferent to the vast majority of regulations that don't really have an effect on them as individuals.

Is it possible that some regulations uniformly benefit all consumers or uniformly harm all consumers? It is possible to construct regulations that serve no purpose and harm everyone, but it is very difficult to make sweeping statements about uniform consumer views on all regulations. There are some groups who claim they speak for all consumers on regulation. But this is impossible: actual consumer views on regulation are as varied as there are many people. Indeed it seems at times that some consumer groups almost get it exactly wrong. 

As an example, about two years ago AEI sponsored a paper here in this very conference room in which Prof. Jerry Houseman from MIT gave an extremely interesting paper on the distribution effects of taxes on various forms of telecommunications services. And the result of Professor Hausman's paper was that placing taxes on the elastically demanded minute-of-use charges on telephone services is one of the most punishing taxes in the arsenal of the American government. It is a tax that probably ought to be gotten rid of. 

In contrast a far more efficient tax would be one that is not user sensitive, just a simply a flat tax on lines.  Consumers would lose far less welfare with that form of tax, but both of the forms of tax are far less efficient than the simple personal income tax that the government administers. You would think that consumer groups would be all in favor of the tax that is the most efficient, the one that harms consumers the least.  But, paradoxically in a proceeding before the Commission right now to reduce access charges on long-distance telephone calls, many consumer groups are opposed to shifting access charges--which are usage-sensitive taxes--to a flat tax. Instead, many of these groups want yet another usage-sensitive tax, the universal service fee. This position is to me an extraordinarily anti-consumer outcome, and one that I find quite paradoxical as being advanced by consumer groups. 

Let me also note that the question of cross-price demand effects on the demand for regulation is interesting. In many cases, consumers do not face other price effects because they have no alternatives. Regulation is a government monopoly and consumers must take or leave it.  There simply  is not much choice of water sources for consumers who want to clean their clothes in their own home. They have to purchase tap water from the local water source.

Consumers may be able to substitute demand for some unregulated services or goods.  For example, Consumers can switch to the Internet for long-distance service. In this case, less Internet regulation substantially lessens demand for consumer use of long-distance services. Conversely, raising Internet regulation would likely increase demand for long-distance services.

Let me illustrate the quality characteristics of the demand for regulation with one of my favorite agencies: your friendly Department of Motor Vehicles. Every adult is familiar with it. It is where you go to get a driver’s license, change title to a car, have your car tested for emissions, and so on. The department handles hundreds of thousands of people every year. It is slow, tedious, frustrating. But in the end, it works.

In their demand for regulation, consumers care about at least seven of the ten characteristics of the Realpolitik of regulation that I introduced earlier. Some, but not all consumers, care about:

a) Written rules

When you go to the Department of Motor Vehicles, most procedures are written down in simple detail. Required documents are written down. Even the questions for a driver's license are written down. There is a written standard for appeal if the rules are not followed by DMV.

b) Predictable processes

Not only are procedures written down at the Department of Motor Vehicles, they are followed in a predictable manner

c) Rational rules

While some people may not think that everything that the Department of Motor Vehicles does is rational, it is difficult to fault them for their processes. If you don't think they are doing their process efficiently, you can complain. Yes, it is true that they cannot always hire more clerks, but they are probably fairly  responsive if you have a useful suggestion

d) Simplicity

You don't need to hire an army of lawyers or accountants to get a driver’s license in the United States.

e) Full disclosure

The DMV does not try to hide from the public its failings. If it is going to take half a day in line, the DMV will tell your that it will take a half a day in line.

f) Efficiency

All right, this is not the strong point of the Department of Motor Vehicles.

Finally, and to me perhaps most importantly,

(g) Similar treatment of similarly situated parties

We don't expect to be treated differently from the person in line in front of us or behind us  in line at the Department of Motor Vehicles. We certainly don't expect to be pulled out of line, put in a separate building, and be told that we will face an entirely different process from everyone else.

Consumers do not seem to care much about whether a regulator is acting according to law, expanding jurisdiction, or following transparent processes.  Just this week a high ranking official at the FCC sent a letter to Congress saying that we are not going to meet a date for a spectrum auction that is required by law.  I'm sure most consumers will have a big yawn. Well, what's the big deal?  The law says "Do it." The agency says: "No we don't want to do it." What's the big deal? 

Nor do consumers particularly care about the expansion of jurisdiction.  A few years ago, the Food and Drug Administration decided it could regulate tobacco. That's not precisely what the law said they could do but they decided they wanted to do it anyway. This would be good.

There really was some consumer resistance. The decision was eventually thrown out by the courts, but, for several years, the FDA asserted it had jurisdiction. 

Nor do consumers particularly care about transparency.  Some government agencies are supposed to have transparent processes; others don't.  The Department of Justice is not supposed to have transparent processes.  No one knows exactly how the Department of Justice’s Antitrust Division is coming to its conclusions about the Microsoft case, but there isn't a profound consumer outrage about that either. 

But we do care about discrimination.  We Americans like to express our views, but we do not necessarily expect anyone to change his or her business behavior as a result of our opinion.  Try this experiment the next time you are at a grocery store.  Tell the cashier your views on the environment, and see if the cashier charges you any more or any less depending on your views about the environment. Of course, there will be no difference in the amount charged. A few comments on the environment may provide a smile, or frown, but little else. We Americans demand and expect to be treated the same as everyone else in line.

Now flip the experiment around  and suppose the cashier lectured you on the environment and then offered you a 10-percent discount if you signed a petition or if you decided not to sign the petition you would be charged 10-percent more, would you be offended?  Of course you would be, and quite possibly you would march into the managers office, and demand that the cashier refrain from these practices or be fired on the spot.  When it comes to business dealings, we Americans expect to be treated the same as anyone else, regardless of policy viewpoints. Businesses that treat their customers or suppliers differently, based on their viewpoint would, likely, soon go out of business.

Expectations of non-discriminatory treatment based on viewpoint extend not only to business dealings but to government activities as well. A parent would be dismayed if the principal of a school set school rules that favored or disfavored children based on the political viewpoint of a child or a parent. A property owner would be offended if the county council set property taxes based on the political view of owners such that identically situated property owners are taxed at different rates.

We Americans are not so naive as to believe that lawmakers and rule makers have no political views. Of course they do, whether they are members of Congress, a State legislature, a county council, or the local school principal. But we do expect the laws and rules that they write to apply equally to equally situated individuals or businesses. And we don't expect laws and rules to single out for reward or punishment individuals based on their viewpoint.

Our expectations of neutrality also extend to those who enforce laws and those who adjudicate disputes. Don't expect a sympathetic ear on environmental policy from a policeman as he writes you a ticket for parking violation.  Nor should you expect the traffic court judge to be amused by a long testimonial about your  good ideas on the environment. 

We would be shocked however if we flipped the experiment. As the policeman is writing you a parking ticket, the policeman said: "You know, this is what I think about the environment. If you agree with me, I won't write the ticket."   Or what if the traffic court judge said: "Your stock of good will in this court has been greatly diminished because you wrote to your Senator to support environmental legislation with which I personally disagree.  Now if you were to withdraw your letter to your Senator, perhaps this parking ticket might go away."  

We Americans expect our government officials -whether federal, state, or local - to treat each of us anonymously. We don't expect our views or our behavior in unrelated matters to be held against us by government officials. 

Now, let me describe the supply of  regulations. How do regulators supply it?  Keep in mind, regulators are monopolists.  In economic terms, they are not price takers like competitive entities. They implicitly set marginal revenue or marginal benefits equal to marginal costs where benefits are unlikely to be revenue, but a motive associated with some welfare function other than profit. Which welfare function?  The public's? The agency's? The individual  regulator's? Who knows? It can and does vary by regulator. Is the welfare function to follow the law?  To promote certain values?  Promote competition? Protect consumers? When you listen to regulators in Washington, you'll hear each of these and other welfare functions described.  But all too often, it is not simply "follow the law"; instead, it's something else. 

Moreover, the marginal costs in the monopoly calculation are not necessarily the costs to a business, industry, or even the economy. They may be some other form of cost, perhaps even the costs of regulation to the regulatory agency.

In any case, we do not have the standard monopoly result for profit maximization for a normal good in which the supply by a monopolist would be less than what exists in the competitive market. If anything, there is far more regulation than exists in some hypothetical competitive market for regulation, although it is difficult to describe what a competitive market for regulation would look like. This empirical result is at least suggestive of the shapes of the regulatory marginal benefit and marginal cost curves in the regulatory decision-making process relative to the actual economic marginal benefit and marginal cost curves.

The actual economic description of prices and technology in the formulation of regulation by government monopolists is a fascinating topic for yet another chapter of this unwritten book on regulation. Suffice it to say, one could take any standard textbook on public finance or on regulation and develop a fairly accurate description of what rational regulation might look like. And at the end of the day, it would likely have many of the quality characteristics that many consumers demand, certainly rationality with  benefits exceed costs.

But for anyone who has spent any time in Washington, one realizes that that is not how regulation is done in Washington. There isn't a textbook model where regulators look at a social welfare function and try to calculate costs and benefits of regulation. It simply doesn't happen.

Moreover, in a textbook setting, one might expect some consistency across agencies and across regulators in the 10 characteristics of Realpolitik that I have discussed. In practice, they diverge substantially. To some extent, they diverge because some regulators have an objective to go out and  simply "do good."  To some regulators, "doing good" is what regulation is all about, and to do the most good you want to do it by having the most discretion which gives you the opportunity to do even more good.  

So how does an agency do the most good? Here are some strategies that are often followed by Washington regulators.

(1) Don’t worry about strictly following the law. Well, frankly, the law is often vague. What do you do if the law is vague and at times even self contradictory? Why follow it if you are not required to do so? When agencies don't follow the law, there's no punishment.  

(2) Expand agency jurisdiction. Never miss an opportunity to expand agency jurisdiction. This goes on at every agency in town  They go into court and assert they have lots of jurisdiction. The next time they're in court, they have even more jurisdiction. For instance, the FCC has held that all packet-switched telecommunications services are interstate in nature and therefore under FCC jurisdiction.  This might put state regulators out of business. It probably extends FCC jurisdiction to central Australia, central Africa, and outer space. There's no limitation on the assertion of FCC jurisdiction. And it is a good way of getting rid of potentially competing regulatory bodies.  

(3) Refuse to write procedural rules. This is one of my favorites.  The FCC has tens of thousands of license transfer applications every year  The FCC has few written rules though on how to deal with these tens of thousand of license transfer applications every year.  If you want to know how you're supposed to be treated, there is no written process for guidance. You hire a Washington lawyer to try to figure it out. At the of the day, you have no written rule on which you can go to the FCC or to court to complain that procedural rights are not being followed. 

(4) Refuse to have transparent processes. An agency gets a bit more discretion if outside people don't know exactly what you're doing.  For example the SBC/Ameritech transfer license of a year ago involved weeks and months of meetings behind closed doors. The public wasn't invited to these meetings. Who knows what happened at these meetings? A deal was struck that involved a lot of conditions on license transfers, conditions that had nothing to do with the transferred licenses. But these conditions had everything to do with unrelated matters involved in the merger. The parties, facing a monopoly regulator, agreed to the conditions for the lack of an alternative.

Many of the conditions broke the law. Sections 251 and Section 252 of the Telecommunications Act of 1996 were violated by the negotiated conditions. The only parties, however, with standing to object apparently were the parties that cut the deal. The FCC has not been held accountable.

(5) Refuse to have predictable processes. At the FCC, timelines and possible conditions for license transfers are unpredictable. If you are one of the tens of thousands of license transferees at the FCC, you really don't where you stand in the process.   The Office of General Counsel has invented some timelines without Commission approval. The same Office of General Counsel can unilaterally change those timelines.

(6) Refuse to collect, disseminate, analyze, or evaluate systematically information on the costs and benefits of regulation.  When I was Chief Economist for the House Commerce Committee, the Committee would send out annual survey forms for dozens of agencies under the Commerce Committee jurisdiction. The surveys had some simple questions about the collection of information on the costs and benefits of rules that the agency had made in the past year. Not a single agency had any complete information for a cost benefit analysis of rules. The vast majority of agencies in Washington do absolutely no cost-benefit analysis, and the few that do any analysis do it only minimally as required. No agency has a clue about whether a single regulation that had promulgated in the past year or any time in its history had benefits that exceeded cost.  

(7) Make regulations complex. Complexity seems to define regulation. At the FCC, we have a computer cost model that is supposed to describe the cost of telephone service. Now, you might think the FCC would know something about the cost of phone service. We regulate a lot of phone service, and we collect much information on those regulated services. We know what rates companies charge customers for a large range of services. If any government agency should be able to point to some cold, hard facts for the costs of phone service, it should be the FCC. And we should be able to describe them in a hurry.

The FCC does not seem to be in much of hurry when it comes to describing the costs of phone service. Nor did we want to make it simple and transparent.  It seems, we needed a cost model that no one could possibly understand, because if people understood it then they might challenge it.  So how do you develop a cost model that no one can understand? You hire a bunch of programmers to develop a model that is not intended to be a cost model. Then you force the new model to be a "cost" model. 

The net result is a cost model that takes 180 hours to run. I don't think you have to be a rocket scientist to know that, in the year 2000, a model that takes 180 hours to run is a model that does not work. By definition, it doesn't work. No one believes it works. Everyone know that it doesn't work, but it is too complicated to say precisely why it doesn't work, because it takes more than a week to do one run. Yet this is the cost model that the FCC insists on using for universal service. The agency would like to use it for all sorts of other proceedings. Few people believe its results, yet the agency hides behind the model’s complexity to be able to say: "It's so complex it must be right."

(8) Be inefficient. We'll leave that for another day.  

(9) Promote benefits while hiding of harm to consumers.  Part of the game of regulation is to persuade people that you doing a lot of good and you're not doing any harm.  So one way of doing that is to prohibit people from describing any harms that there might result from regulation. For instance, last year the FCC had a proceeding that is called the "Truth and Billing" proceeding.  Now, anytime any government agency employs the word truth, watch out! It is a word that is more commonly used by other governments-like Cuba or China. It is word rarely used by the government of the U.S.

This proceeding essentially prohibited telephone companies from describing the details of universal service programs to the consumers.  Companies are prohibited by regulation, not by statute, from telling their consumers about the costs of universal service. The public doesn't know.  Cal Thomas wrote an op-ed piece earlier this week in which he praised Bell Atlantic for disclosing the amount of his universal service fee contribution. But Cal Thomas wasn't really told for what purpose the money was collected, and he wasn't told precisely how much Bell Atlantic was contributing on his behalf.  

(10) Treat similarly-situated parties in different ways. This is the fundamental basis for the realpolitik of regulation.  For maximum discretion, a regulator must treat different parties in different ways. How do you do this? Well, one of the ways you do this is you promote the idea that there is such a thing as good will, or political capital at agency.  Recently a high-ranking official at FCC suggested that a certain industry had  lost "good will" at the FCC by lobbying for legislation at Capitol Hill. Are industries or businesses treated differently at regulatory agencies depending on their stock of good will at any given point in time? Do agencies keep running tabs on good will?  Does an outside party lose its free speech and opportunity to petition Congress if it is involved in regulatory proceeding?   The answers to these questions are just too chilling.

With all of these ten attributes of the realpolitik of regulation at work, it is impossible to determine whether the benefits of regulation exceed the costs. It is not surprising, therefore, that agencies refuse to engage in the collection, dissemination, and analysis of the cost and benefits of their regulations.  Different agencies have different combinations of these characteristics. In general we are at a disequilibrium where consumer demand does not match the regulatory supply, because regulators often have no incentive to do so.  Perhaps the regulators really are god.  The offensive strategy for regulators in the realpolitik of regulation is to expand these characteristics; the defensive strategy for everyone else is to diminish them. Although a regulator by trade, I am interested in diminishing these attributes at the FCC and  at other agencies.