The Game of Telephone: The Knowledge Problem in Regulation

Why do economists who accept a theory oppose putting it into practice?  For example, I believe global warming is a rather significant problem.  I agree that it is internally consistent that carbon taxes (or some other variation like cap & trade) can reduce carbon emissions to a socially optimal level.  So, why then do I oppose carbon tax regulation?

There are many reasons why I (and many other GMU-style economists) oppose regulation even though a logical argument can be made, it could improve a given situation.  We tend to focus on public choice reasons (such as rent-seeking and agency capture).  The knowledge problem, most famously discussed by F.A. Hayek, is also often cited: government agents can too seldom possess all information and knowledge necessary to regulate desirably and much less “optimally.”

There is an element of the knowledge problem that warrants further attention, an element highlighted by Don Lavoie in his 1985 book National Economic Planning: What is Left? In this book, Lavoie greatly expands our understanding of the knowledge problem and its relevance for assessing central planning and more mundane government regulation.  He discusses Hayek’s formulation of knowledge as mostly tacit, but Lavoie also emphasizes that knowledge is built upon inarticulable foundations.  Attempts to articulate the inarticulate foundations are doomed to fail as each person carries with him or her different nuanced understandings of the language used in legislation authorizing regulations.

Consider, for example, the phrase “2+2=4.”  Understanding the phrase’s meaning requires a tacit, inarticulable understanding of the elements: 2, +, =, and 4.  If one were to try to rigorously define every element in that phrase, he would eventually fall into a problem of recursivity.  As children, when we first encounter mathematics, it may seem weird and arbitrary.  We just learn that 2+2=4 by rote.  It is only through repeated interactions with mathematics do we start to understand it.  To paraphrase the great mathematician John von Neumann, you never really learn mathematics.  You just get used to it.

The problem of inarticulable understandings of knowledge comes into play in the field of regulation.  The economist has a foundation of knowledge.  When he tries to convert that knowledge into policy, we run into a game of telephone.  At each step along the way, the knowledge and information get a little distorted. Each person has different foundations from which they understand the message the economist is delivering.  As such, the end policy would deviate considerably from theory, even if we assume away public choice issues.  In other words, the policy will look considerably different from the theory because of a sort-of language barrier.

Consider, for example, the word “cost” in economics.  We define “cost” to mean what one gives up to take a particular action (it is sometimes called “opportunity cost” for this reason).  Cost is inseparable from choice.  Yet, “cost” takes on a very different meaning for the general public, as it usually refers to a negative consequence (“the cost of reading is a headache”) or the monetary price of something (“the coffee cost me $2”).  Thus, the economist already faces a problem communicating his theory to policymakers.  But even within the field of economics, “cost” has different understandings.  James Buchanan’s excellent short 1969 book, Cost and Choice, discusses how “cost” has changed understandings among the various schools of thought.

I oppose regulation even when I understand the argument because argument and policy are not the same things.  When communicating, experts run into the telephone problem: the theory is misunderstood, misapplied, or miscommunicated.  Competition among experts helps solve these problems, yes, as experts become incentivized to be less wizardly and more like teachers.  But the knowledge problem remains, and regulation can only enhance the communication problems.


Read More

Will Joe Biden Be a Dictator?

This might look like a ridiculous question to ask about a soft-looking near-octogenarian who signals his virtue by repeating the inclusiveness mantra. But not so much if you define “dictator” as a political ruler who imposes on the whole population some shared preferences of the minority who brought him or keeps him in power. A more inclusive definition would replace “minority” by “majority short of unanimity.”

Biden was elected by 51% of the American voters. If, to be inclusive indeed, we include the third of the electorate (that is, of Americans eligible to vote) who did not vote, Mr. Biden’s support shrinks to 34% (51% × 66%). Now, consider that many who voted for him probably did so only or mainly because they thought that his adversary, Donald Trump, was even worse—not an unrealistic hypothesis. If Biden imposes the preferences of 17% of the electorate to 83%, or even of 34% on 66%, he will be a dictator. (Note that my definition of the term is not very different from the one in Kenneth Arrow’s famous Impossibility Theorem.)

An interesting article that bears on this topic is John G. Grove’s “Numerical Democracy or Constitutional Reality,” Law & Liberty (our sister website), November 12, 2020. Grove argues that the United States is a limited, compound republic, not a numerical democracy, and that the whole check and balance structure is meant to prevent a numerical majority from bulldozing the preferences of the rest. In this perspective, each side has a right to have its preferences incorporated in the winner’s legislation and an adverse electoral result is not, for the losers, a catastrophe to be corrected at any cost.

By the very nature of government, however, it is not easy to prevent winner-win-all results: a law is enforced against everybody, especially against individuals who did not agree with it. It seems that, on the basis of an individualist philosophy, only a near-universal consensus could justify radical change.

One disturbing implication is the following. Grove’s idea is a two-edged sword. When we start from several decades of a collectivist legislative and regulatory drift that has trammeled the minority of individuals who want to be largely left alone, even a new numerical majority may not and could not rapidly change course. Ronald Reagan, with his many good ideas (and a number of bad ones) did not bring much change and perhaps no lasting change. But for the same reason, thank God, Trump was not able to do more damage than he did.

James Buchanan, the Nobel economist, understood the conundrum: How can one reverse dictatorship without being a dictator himself? The solution, Buchanan argues with Geoffrey Brennan in their book The Reason of Rules: Constitutional Political Economy (Liberty Fund, 2000[1985]), is a “constitutional revolution.” That is, we—“we” classical liberals and libertarians—need to promote radical change to which our fellow citizens can unanimously consent, at least in theory. This pedagogical and abstract task is not an easy one.


Read More

Margins and the 2020 Presidential Election

The Power of Thinking on the Margin

Because I understand the power of one vote–it’s very close to zero–I always vote in Presidential elections for the candidate who’s closest to my views. The first time I was able to vote in a Presidential election was in 1988 and from then until now I have voted for the Libertarian Party candidate.

That’s where thinking on the margin has led me.

But Presidential candidates have a much thicker margin. They make hundreds of decisions–about where to speak, how to debate, and what to say. When they are incumbents, they have a large input on many policy issues that can affect the outcome of the election.

Health economist (and friend) John C. Goodman sent me an email Monday with the provocative title “Why Trump lost the Election: Health Care.”

In it, he writes:

The editors of the Wall Street Journal, the editor of National Review (Rich Lowry) and John Goodman all agree: Trump didn’t endorse the plan outlined by Goodman and Heritage Foundation scholar, Marie Fishpaw.


Trump actually did the things Goodman and Fishpaw recommended, including allowing people to talk to their doctors by phone, email, and Skype; allowing employees to have access to 24/7 primary care as an alternative to the emergency room; and allowing employer-provided health insurance to be personal and portable. But Trump never talked about any of this. So, he didn’t get credit for any of it.


I think John is right. But one could also say that if he hadn’t been so incredibly rude and nasty in the first debate, he would have won also. (Although we now know in retrospect that Trump was probably awfully sick with COVID-19 during that first debate. When you’re sick, you tend to let out your inner self. And Trump’s inner self is nasty.)

Consider the fact that if Trump had received just 43,000 more votes, properly distributed, in Arizona, Georgia, and Wisconsin, he would have won the electoral votes of those three states. That’s a total of 37 electoral votes. Had Trump won those, he would have had 269 to Biden’s, wait for it, 269.

What would have happened then? It would have been thrown into the House of Representatives where each state delegation gets one vote. So California gets one vote and Rhode Island and Montana each get one vote. Etc. The vote is based on the November 2020 election results. Based on those results, Republicans had 26 votes. In that case, Trump would have won.

Interestingly, though, he might have had Kamala Harris as his Veep because the vote for Veep would have been by U.S. Senators. This is unclear, though, because the Senate is tied 50-50. Does anyone who reads this know?

Now back to the main point: Trump’s thick margin. As Holman Jenkins pointed out in an aptly titled Wall Street Journal opinion piece, “Trump Threw it Away,” January 6, 2020, Trump almost won. Jenkins wrote:

Of course the microscopic margin rankles—he lost the pivotal electoral votes of Georgia, Arizona and Wisconsin by fewer than 43,000 votes. He has every reason to be beside himself since he absurdly oversupplied these voters with reasons to vote against him and he still almost won.

Imagine a team so bad and good at the same time that it would have prevailed if it had fumbled the ball 1% fewer times in its own endzone.

Imagine what would have happened if Trump had been neutral, not nice but simply neutral, to the memory of John McCain. He probably would have won Arizona. (Of course, that’s like asking what would have happened if Trump hadn’t been Trump.) What if he had pointed out the record growth in median incomes for various minority groups? He might have won Georgia. What if had actually run a campaign based on his accomplishments up to the end of 2019? He might have won Wisconsin. Etc.

So although we voters can’t individually affect the outcome, candidates can influence the outcome with a few key decisions.


Read More

A Shocking USPS Admission

USPS: We Don’t Care; We Don’t Have To

Last Monday, January 11, I mailed off my estimated tax payments to California’s state government (an agency called the Franchise Tax Board) and the federal government (IRS). Both, but especially the check to the IRS, were for large amounts.

At about 10 a.m., I put them in a mail box close to my office in downtown Monterey. The sticker on the box says that pick up is at 12 noon.

At about 2:20 p.m. I received a call from a man at a nearby men’s club. He explained that he had received my letter to the Franchise Tax Board in that day’s mail. He had used my return address to track down my phone number. I thanked him profusely, rushed over to the club, and picked up the letter. My concern, of course, was about what happened to the check to the IRS. Was it badly misplaced also? And if it was, would the person who received it be as responsible and as resourceful as this nice man?

I walked from there to the downtown post office. When it was my turn, the woman I talked to said she would get the postmaster. I waited about 10 minutes until he came to the front. I started to explain the situation. My first surprise was that he told me that the mail from that box hadn’t been picked up yet. I was surprised because I was showing him the letter that I had put in that box. How did he think I got it? When I finally got him to listen and understand, I told him my bigger concern: was the IRS check similarly misplaced?

He answered that he didn’t know and couldn’t know at this point. Then he told me a shocking number. He said that the postal service loses about 2 percent of the mail. Even though I had my mask on, I think he saw my eyes widen.

“Two percent?” I said, “that’s terrible.”

“No, it’s not,” he answered. “We have billions of pieces per day.” (I think he overestimated by at least an order of magnitude, but that doesn’t matter for these purposes.)

“Two percent is a large number,” I answered. “FedEx doesn’t lose 2 percent of its shipments.”

“You can’t mail something with FedEx,” he said.

“Yes, you can,” I replied, “You can put a letter in a FedEx envelope.”

“Then use FedEx,” he answered.

“Meanwhile,” I said, “I’ve got this problem that I did use USPS and I want to make sure my check to the IRS will be delivered.”

“We can’t cross that bridge until we come to it,” he said.

I didn’t understand what that meant in this context, so I asked, “What’s that bridge look like? How would you know we’ve come to it?”

“We would know when the IRS contacts you and says they didn’t get the check.”

“I’m pretty sure they don’t get in touch when that happens,” I replied. “I won’t know until April when I file my taxes and they tell me I owe a big amount rather than sending me my usual small refund. They would also charge me interest and a penalty.”

“So that’s how we would know,” he said.

(I’ve since realized that if I don’t see that the check has cleared by, say, January 22, then it was lost. It’s not clear what I would do. Would I call the IRS? Good luck with that. Would I put a stop payment on the check and send another one? I’m nervous about putting a stop payment on a check to the IRS.)

One of the interesting things about the interaction, which shouldn’t have been surprising, is how little this guy seemed to care and how blasé he was about a 2% loss in mail.

As I say, though, incentives matter. And the USPS’s incentive to deliver all the mail is very low.


Read More

Ralph K. Winter Jr. RIP

Catching up on Wall Street Journals from December today, I came across an obit of federal judge Ralph K. Winter. I never met the man although I gather that a number of my friends have. But it’s amazing how one quote can stick out in your memory from over 40 years ago. I remembered that quote and found the publication it was in.

Winter wrote “Campaign Financing and Political Freedom” for the American Enterprise Institute in October 1973. I think I was on the mailing list for their domestic policy publications. (I think my marked up copy disappeared in my fire.)  One reason those pubs were really good is that Yale Brozen, an economics professor at the University of Chicago, had a real academic entrepreneurial eye for a good study and had a large role in choosing authors and topics. (Incidentally it was written with John R. Bolton, who became famous for his hawkish foreign policy views.)

I remembered that Winter had made a strong cogent case against the campaign finance laws that were about to happen and I found it persuasive.

Here’s the part I remembered clearly and still love:

Candidates seem never to lose because the public is indifferent to them or to their platforms; they seem to lose because they cannot raise enough money. Tom Wicker tells us that Fred Harris and Paul McC!oskey saw their campaigns founder “for want of means to wage a primary campaign,”” a state­ ment that is true in the same sense that if a mayoral candidate in New York City were exposed as Martin Bormann, his withdrawal statement would mention only difficulties in raising campaign funds.



Read More

Newt Gingrich’s Numeracy Problem

And Trump’s and the Dems’ arithmetic problem.

$2,000 * 200 million does not = $2,000.

$2,000 – $600 does not equal $2,000.

Newt Gingrich tweets:

If Senate Republicans fail to bring up the $2000 payment as a clean vote they run a real risk of losing the two seats in Georgia. This is an 80% issue. People get it. Billions for the banks, billions for big companies, but we can’t find $2000 for everyday Americans.

If the proposal before the Senate really were to give $2,000 to everyday Americans, no one would be raising an objection because $2,000 divided by, say, 200 million everyday eligible Americans is way, way below 1 penny each.

Everyone understands that it’s $2,000 per “everyday American.” With about 200 million Americans qualifying, that’s $400 billion.

So if we were to rewrite Newt’s tweet honestly and accurately, it would read something like:

If Senate Republicans fail to bring up the $2,000 payment as a clean vote they run a real risk of losing the two seats in Georgia. This is an 80% issue. People get it. Billions for the banks, billions for big companies, but we can’t find $400 billion for everyday Americans.

Sounds a little different, doesn’t it?

And remember how Trump objected that $600 was way too low an amount to have the government give eligible people and $2,000 was the right amount? Well, they got the $600. So if Trump really believed what he said, and if the Democrats believed what they said they believed, he and they should have pushed for an extra $1,400, not an extra $2,000.

Newt’s tweet also shows the difference between those of us who want to have gridlock and divided government in order to restrain government and people who want divided government simply because they want the Republicans to be the majority party in the Senate.

Many of us want gridlock because we fear what a Democratic House of Representatives, a Democratic Senate, and a Democratic president will do. One of the main things many of us fear is that they will spend hundreds of billions of dollars more than if the Republicans won the Senate and restrained the Democrats’ spending. But if Mitch McConnell caves so that the feds spend an extra $400 billion and the Republicans win in Georgia, they will have nullified a huge part of the reason for having the Republicans win in Georgia.


Read More

Joe Stiglitz on Taxes

Taxation is unlike most transfers of money from one individual to another: while most other transfers are entered into voluntarily, taxation is compulsory. In Chapter 5 we saw some of the reasons why the contributions to support public services need to be compulsory: because of the free rider problem, unless support for public goods is made compulsory no one will have an incentive to contribute. We showed, in particular, that all individuals might be made better off by voluntarily agreeing to be compelled to contribute to the support of public goods. Yet the ability to compel individuals to contribute to the support of public goods may also provide the government with the ability to compel individuals to contribute to support some special-interest group: the government has the power to force one group to give up its resources to another group. This forced transfer has been likened to theft, with one major difference: while both are involuntary transfers, transfers through the government wear the mantle of legality and respectability conferred upon them by the political process. In some countries and at some times, the distinction becomes, at best, blurred. The political process becomes detached from the citizenry and is used to transfer resources to the groups in power.

This is from Joseph E. Stiglitz, Economics of the Public Sector, Second Edition, 1988. I used the textbook in a policy analysis course at the Naval Postgraduate School in the mid to late 1990s.

I find the order of his reasoning in the above fascinating.

The first sentence makes clear, in no uncertain terms, that taxation is compulsory.

The second sentence is the typical case for taxation to finance public goods, something that he might have learned (if he hadn’t already known and he probably did know) at the knee of his mentor, the late Paul Samuelson. It also contains the standard exaggeration: no one would have the incentive to contribute? I don’t know of any such examples. Even for the economist’s typical example of national defense, if some foreign invader is coming at me, I’m going to defend myself. Is the amount of defense enough? Probably not. But my incentive is not zero.

The third sentence reminds me of James Buchanan when he channeled Knut Wicksell. Buchanan took seriously Wicksell’s idea that the only way to be sure everyone benefited from government spending is to require unanimous consent for every tax/spending program. Of course, in practice this is unworkable, but it’s a good reminder.

The fourth sentence lays out the huge downside of relying on taxes: taxes can be used to redistribute wealth. Indeed, that is the main thing they do in most countries. Here Stiglitz is channelling Public Choice theory.

In the fifth through seventh sentences, Stiglitz is way more radical in an anti-government direction than I ever would have expected.

There’s so much I like about this textbook. I don’t think he would write it even close to the same way today.





Read More

Did the Libertarian Party Cost Donald Trump the Election?

No, but it might have cost him Georgia’s electoral votes.

My friend and fellow economist Walter Block has an op/ed in the Wall Street Journal (November 8 and November 9 print edition) titled “Libertarians Spoil the Election.”

Here’s his argument:

Did the Libertarian Party throw the election to Joe Biden? Maybe. At this writing nominee Jo Jorgensen’s vote total exceeds Mr. Biden’s margin over President Trump in Arizona, Georgia, Nevada and Pennsylvania, enough to change the outcome.

First, he’s wrong about Pennsylvania and Nevada. Jorgenson’s vote doesn’t cover the spread.

He has a better case for Arizona and Georgia.

But even there, here’s the problem: Walter is assuming implicitly that the vast majority of votes that went to Jo Jorgenson would have gone to Trump. I think that’s wrong for two reasons.

First, I would bet that about 20 percent of the people who voted for Jorgenson would not have bothered voting had they not been offered that alternative. (What’s my evidence? I admit that it’s gut feel.)

Second, consider the remaining 80 percent. I would bet that at most 2/3 of this remaining 80 percent would have voted for Trump had Jorgenson not run. Why as much as 2/3? Because what I have observed is that young libertarianish people would have preferred Biden over Trump and older libertarianish people would have preferred Trump over Biden, and a much higher percent of older people than of younger people vote.

If I’m right, that means that we would have to take the difference between 2/3 of 80% and 1/3 of 80%, which is, of course 1/3 of 80% and apply that to the Jorgenson totals in each state.

Do that and Walter’s point might work for Arizona and Georgia but it’s not a slam dunk.

Arizona: Biden gets 1,645,277 votes, Trump gets 1,629,845 votes, and Jorgenson gets 50, 121 votes.

80% of the Jorgenson vote = 40,097 votes.

1/3 of that = 13,366 votes.

Biden minus Trump = 15,432.

So even there, not clear that Trump would have won Arizona.

Georgia: Biden gets 2,467,870 votes, Trump gets 2,456,275 votes, and Jorgenson gets 61,951 votes.

80% of the Jorgenson vote = 49,561 votes.

1/3 of that = 16,520 votes.

Biden minus Trump = 11,595 votes.

So there there’s a much better shot at Walter’s point.

In his op/ed, Walter makes a strong case for Trump over Biden, most of which I agree with. Walter is critical of Trump on protectionism, as he should be. But he does leave out a major issue, one on which Biden is head and shoulders above Trump: immigration.




Read More

Voting and Margins

I used to teach my students, before I was allowed to vote in this country (I became a U.S. citizen in 1986) that even in swing states, their vote for President would not be determinative.

When I finally got to vote (I think it was in the June 1986 California primaries), I voted and experienced the truth of my statements.

My vote for U.S. president makes no difference on the margin and neither does yours. (That’s why I always vote for the person closest to my views, no matter how slim his or her odds.)

Which, of course, doesn’t mean that politicians running for office should think that votes don’t matter. They’re dealing with much thicker margins.

My favorite example I liked to give in class after 2000 was the Bush/Gore election for President because all my students had followed it, at least somewhat. One of the big issues in Florida was that, on April 22, 2000, Bill Clinton’s administration had used guns to rip away Elian Gonzalez from his relatives in Florida and send him back to Cuba.

That was a big upset to a large number of Florida voters from Cuba or whose parents left Cuba. Gore, of course, was part of the Clinton administration. So what was he to do? I followed it pretty carefully and my recall is that Gore lamely criticized Bill Clinton for one news cycle and then let it drop.

Then I asked my students: What if Al Gore had lambasted Clinton for it over, say, 3 days? Is it conceivable that he would have shifted, say, 0.2 percent of the Florida Cuban vote? If so, we would be referring to President Gore.

Or, I pointed out, George W. Bush was an effective campaigner. What if, instead of thinking he had Florida in the bag, partly because his brother was governor, he hadn’t gone home to Texas to do a premature victory lap? Then we might have avoided the legal nightmare of Bush v. Gore.


Read More

From Democracy to Populist Rallies

In his 1945 book On Power, Bertrand de Jouvenel wrote:

Democracy, then, in the centralizing, pattern-making, absolutist shape which we have given to it is, it is clear, the time of tyranny’s incubation.

Sometimes, democracy in America looks a bit like the South American version.

In general, democracy as we know it works differently than what Christopher Achen and Larry Bartels call its “folk theory” version. In their book Democracy for Realists, they describe the folk conception of democracy:

In the conventional view, democracy begins with the voters. Ordinary people have references about what their government should do. They choose leaders who will do those things, or they enact their preferences directly in referendums. In either case, what the majority wants becomes government policy. … Democracy makes the people the rulers, and legitimacy derives from their consent.

That nothing like this happens in the real world should be obvious by observing the current electoral campaign in America, reading the political advertisements, and listening to the presidential candidates of the two main political parties. Public choice theory explains many failures of folk democracy.

There is much to disagree with in the alternative proposed by Achen and Bartels, which is an elitist democracy based on group identities. But the populist solution does not produce better democracy. It is more an extreme form of “folk theory” democracy that worsens democratic failures.

What is said and done in President Trump’s rallies is more entertainment than information on a political program. These rallies are liturgical spectacles of fusion between the great leader and “the people.” Extreme illustrations were given when the president danced as the crowd was chanting “YMCA” (videos are available on YouTube).

This sort of show is not new in contemporary populism. In his book Populisms: A Quick Immersion (which I review in the forthcoming issue of Regulation), Carlos de la Torre, a professor of sociology at the University of Kentucky, observes that “[p]opulism blurs the line between politics and entertainment.” Of Rafael Correa, the populist president of Ecuador from 2007 to 2017, Torre writes:

Rafael Correa’s campaign strategy in 2006 was also based on mass rallies, where common people were in close proximity to the candidate and sang along with him to revolutionary music of the 1960s and 1970s. Even though this music was retro, Correa’s political rhetoric was innovative. Unlike the long and boring speeches of his rivals, Correa blended music and dance with speech-making. He spoke briefly, presenting a simple idea, music was played, and Correa and the crowd sang along the campaign tunes and dance.


Read More