Five Arguments and a Funeral

I just started reading Ed Feser’s Five Proofs for the Existence of God. It is what I expected it to be, and I’m going to critique each of the five arguments because I don’t think they are very good arguments, certainly not rising to the level of ‘proof.’

Argument vs. Proof

Well, what is a proof? A proof is a demonstration of the connection between a logical or mathematical statement on one hand, and a set of axioms, on the other hand, using only an accepted set of logical, inferential operations.

Proofs of God’s existence usually trade on sloppy definitions, imprecise arguments, and equivocation. All of this is easy in arguments made in human language, but it falls apart when you try to recast the same arguments in strictly logical terms.

But more fundamentally, is God a theorem or an axiom?

Axioms are the ideas we have to take for granted. From them, we can derive theorems, which depend completely on our choice of axioms. And it is a choice, we can use any set of axioms we want. For example, in classical geometry, there is an axiom known as the Parallel Postulate. One way to state the Parallel Postulate is that parallel lines never meet even if extended to infinity. Another way to put it, that avoids the concept of infinity, is that the number of degrees in the sum of the interior angles of a triangle is exactly 180.

Because the Parallel Postulate was cumbersome to state, considerable effort was expended to show that it wasn’t an axiom – that it could be derived from the other axioms of Euclidean geometry. However, mathematicians in the early 19th century showed that it really was an axiom. By changing the axiom to versions where the interior angles added up to more or less than 180 degrees, entirely new geometries were invented. In particular, spherical geometry uses the greater than 180 assumption, and is necessary to do accurate surveying on the surface of the Earth.

The point here is that what you really are expressing faith in is your axioms, not your theorems. So if you really can ‘prove’ God exists under some system of axioms, you aren’t expressing faith in God, you’re expressing faith in something else.

And BTW, if your set of axioms contains a contradiction, you can prove anything.

So when critiquing these arguments I’ll focus much more on critiquing the axioms, rather than the derived ideas.


One of those key assumptions is that infinite sequences can’t be real. This idea appears several times in arguments that there must be a First Mover, a First Cause because an infinite sequence just can’t be accepted as a possibility. If we eliminate an infinite regress, there must be a First, right? Not so fast.

Imagine you are a tiny, point-like creature that lives on a circle. You want to back up to where you were before. And again, And again. Are you ever going to finish backing up? Obviously not, because you live on a circle, you will eventually back up completely around the circle and keep going around and around. While the circle is finite, it is also unbounded.

“Finite, but unbounded” is just one way of avoiding the need for a First. Another is to point out that we do experience ‘actual infinites’ all the time. Ancient philosophers may have had trouble with this idea but subsequent generations have given us better tools for thinking.

If you took calculus in school, you know it works by taking a limit as something goes to infinity. Taylor series and Fourier series both work by summing an infinite series of terms, which can converge to a finite, useful number. For example, a square wave is the sum of an infinite number of sine waves.

A deeper, physical example is the Feynman integral way of understanding quantum mechanics. In this understanding, particles don’t move along a single straight line, they move along an infinite number of paths simultaneously. The sum of all these motions is the appearance of a straight line in simple cases, but can explain more complex cases.

We can also consider the path of a photon as it moves outward. A ray, geometrically, is infinite in one direction. There was a time in the early history of the universe when it became optically transparent for the first time. At the moment a huge amount of the mass/energy of the universe which was photons escaped outwards at the speed of light. Nothing will ever absorb these photons. They will speed onwards forever.

You might object the future is not as real as the past. While it certainly is not as well known, the broad outline is well enough known to say that if the universe expands infinitely, these photons will also last forever. Photons don’t decay into other particles.

Infinite theorems, unprovably true

Mathematical development since the end of the 19th century and into the mid-20th century led to a confluence of the ideas of proof and infinity. We discovered that any logical system complex enough to support the development of arithmetic also led to an odd circumstance. There would be theorems in the system that were true, but unprovable within that system. If you promoted them from theorems to axioms there would still be more theorems that were true but unprovable. For more on this fascinating topic, look into Gödel’s incompleteness theorems.

Now any explanation of physics is going to satisfy this condition – it is a logical system that is strong enough to include arithmetic. Therefore we can expect that there are going to be corners of physics that also fall into the category of true, but unprovable.

And of course there is no way to know if any particular idea is unprovable or not!


Some readers might ask why I am getting into this subject here on Rational Exuberance. I’ve previously taken on books like this on another blog. But I want to do this here because there is a connection between this kind of thinking and the political issues that are more commonly discussed here.







Stiglitz and Norvig and Gini, oh my!

Executive Summary: A simple model shows that we can achieve positive changes in wealth inequality by returning to a more progressive tax structure.

Joseph Stiglitz has written an important article on inequality in the American economy. One of his key points is that the economy is not inevitable. We create the rules, the framework in which individual transactions take place.

Peter Norvig is a leading computer scientist, and currently Director of Research at Google. As a teaching tool, he created a simple model economy in Python. The Jupyter notebook explains the code and has some very interesting charts. A quick read of the notebook would give the impression that just about every choice of beginning income distribution, interaction, and transaction leads to serious inequality as measured by the Gini coefficient. This blog will show that that conclusion is wrong.

All models are wrong, but some are useful. – George Box

“All models are wrong…” is the statistician’s version of the aphorism “the map is not the territory.” In the case of Norvig’s Economics Game, the fact that it is implemented in Python gives us unusual insight into how the model is wrong and where we can tinker with it.

The key to Norvig’s model is the basic transaction. In this transaction, two individuals pool their resources and roll the dice on how to split them up.

This model of the economy is almost exactly the opposite of most models. If you look at most ‘rational man’ or ‘economic man’ models, they start with a market in which perfectly rational actors, with complete information, transact simultaneously in an infinite number of frictionless transactions that establish the clearing price for whatever goods are being traded. Instead, this model pair up individuals in a high-risk gamble of no information. There is no market, no good being traded, and no price being set.

No bad actors

In addition, cursory analysis of the transaction shows why it creates inequality and why no rational actor would choose to participate. Consider the case of two actors who put in nearly the same amount to the pool, say 60 and 40 simoleons. If the random number generator spits out a split between .4 and .6 the result will be that the actors have more equal amounts of money than when they started.

Considering the number line between 0 and 1, .4 to .6 is only 20% of the line. Therefore, the vastly more probable result, 80%, is that the wealth shares will be more unequal than previously.

The only group for whom the transaction is actually attractive is the very poor.  In this case, the poor actor has very little to lose and only a small chance of losing it, but they might become instantly rich.

Norvig’s results show exactly this kind of churn. Actors gain and lose fortunes in a single transaction. It is all very chaotic. But is it a bad model?

Actually, it is a very good model of the irrational choices people make every day. It is a good model of the substantial risks a farmer takes growing crops, or a business takes creating a product. The model captures perfectly our willingness to make stupid choices by distilling the economy down to the stupidest transaction possible.

So instead of controlling the economy, and assuming everyone only does smart things, let’s see what can be done while assuming the worst. Is there anything that can save this economy from falling into stark wealth inequality?

Yes, there is.

Taxes to the rescue

One of the modifications that Norvig shows is a primitive tax scheme.

In [20]:
def redistribute(A, B, rate=0.31):
    "Tax both parties at rate; split the tax revenue evenly, and randomly split the rest."


flat tax model

Here, we see that a ‘flat tax’ does help some, but as Norvig says

Another surprise: This transaction function does indeed lead to less inequality than split_randomly or winner_take_all, but surprisingly (to me) it still increases inequality compared to the initial (Gaussian) population.

Well, we can improve on that result, even though Norvig does not pursue it. I took Norvig’s Python code and added some functions to calculate a graduated tax and a split that gave 2/3 of the tax receipts to the poorer of the two parties. I also added data for tax rates of 2018 and 1979, the year before Reagan started chopping away at the progressive tax system.

My results show that the current rates, like the flat tax tested by Norvig, don’t help wealth inequality all that much. However, with the progressive rates of  1979, we can actually achieve a very substantial result, lowering inequality to levels of European social democracies.


This is the important point. Even allowing people the freedom to participate in the worst (riskiest and self-destructive) underlying economic transactions, a progressive tax and wealth redistribution system can preserve societal structures at acceptable levels.

As Stiglitz said, we make the rules. We had rules that worked to keep a lid on inequality, and we foolishly blew them up. Lowering top tax rates has never grown the economy, all it has done is make rich people richer. Returning to a more progressive system of taxation and wealth redistribution will benefit our society.

Nerdy postscript

Norvig says he is surprised that a flat tax can’t return the wealth distribution to the beginning ‘bell curve’ distribution. But looking at the Gini coefficients of real economies, that isn’t a realistic goal or expectation.

Norvig’s own ending is important. Data modeling is far more accessible and easier to change than a pen-and-paper Markov model or an R model.

The latest version of Norvig’s notebook (linked above) includes another simple model that is less illuminating of economics but does show that Gini is not necessarily a perfect measure of inequality. As that model shows, it is possible to have a few superwealthy people throw off the calculation of inequality. This shows that multiple measures – of inequality, poverty, and mobility, for example, are necessary for any deeper model of the structural problems of a real economy.

Arthur Laffer vs. The Beatles

The Laffer Curve has been a favorite bit of Republican pixie dust since the Reagan years, the core idea of “voodoo economics.” So simple that the idea fits, famously, on a cocktail napkin. If the tax rate is zero, the government receives no income. If the tax rate is 100%, the government also receives no income because no one is willing to work for nothing. Therefore, the idea goes, there must be a curve in-between zero and 100 that contains a point where government revenue is maximized.

The pixie dust part comes from claiming that current tax policy puts us on the side of the curve where reducing rates will actually increase revenue. That’s where The Beatles come in.

Taxman” is a tune that shares the feelings of the high earning musicians when they learn that they are paying a marginal tax rate of 95% on their income. “There’s one for you, nineteen for me,” says the taxman.

And what happened then? Was that the point when The Beatles downed tools and went on strike for a lower tax rate? No. They just grumbled and kept on making hit music.

Nobody stops working because they pay a lot of taxes. That is the craziness of the Laffer Curve. 95% rates aren’t enough to stop people from working. During wartime, top tax rates in Britain went to 98% and people still worked hard.

So by the argument of the Laffer Curve itself, we now know which side of the curve we are on. If we want to maximize revenue, raise taxes.

Of course, the government is not in the tax-farming business. We don’t want a revenue-maximizing rate. We want a rate that allows us to pay for the government services we want and to pay off previous debts. That rate is still higher than what we have today. And it is a progressive rate, where very rich people pay a lot in taxes.

(Yes, there is a difference between the marginal rate and the actual, blended rate. But for those making millions of dollars a year, the difference is small and not one that invalidates the argument against the Laffer Curve.)


Follow the Money

Where does a corporate tax cut go? Let’s follow one billion dollars as it gets diverted from the IRS into the financial system.

Megacorp gets a billion dollar tax cut. However, on the same day as the money arrives there are no billion-dollar opportunities waiting for investment. Company management invests in new product lines and billion dollar capital investments such as factories with an outlook and schedule of years of planning. A tax cut could be taken away by the next administration – it isn’t a reliable source of funding.

It isn’t reliable, but it is a windfall. Found money goes into quick return financial shenanigans such as stock buybacks and retiring corporate debt. This what happened to the Bush tax cuts in 2004.

OK, Megacorp spends 800 million on stock repurchase and 200 million on retiring corporate debt. Where does the money go?

Most common stock is held by funds – retirement funds, pension funds, and insurance funds. Most corporate debt is held by similar funds. So almost the entire billion dollars is injected as cash into the funds. Funds have rules for what to do with the cash from non-recurring sources – reinvest it. (The same for recurring sources such as dividends and interest payments. The point is that it won’t be a direct or immediate distribution.) So one billion dollars is almost immediately invested into the financial markets.

The money will eventually find its way out of the funds in the form of distributions to individuals at retirement. At this point, many years after it was given to Megacorp, it starts to turn into consumer spending that benefits the pensioners and retirees and the larger economy. During that time inflation and management fees have cut into the amount, though general market growth may have offset that.

And, of course, it will be taxed. But the tax will be paid by a retiree, not by Megacorp.

Bottom line: Corporate investment uses reliable funding sources and has a long timeline. Tax cuts don’t fit that description. The money does eventually trickle into the consumer economy years afterward. Infrastructure investment would create more jobs, immediately, than any tax cut.

Mitigating Political Ignorance

Ilya Somin’s Democracy and Political Ignorance paints a bleak picture of the state of our nation’s political knowledge, and of what can be done to improve it. After having devoted several posts to Political Ignorance, I would be remiss if I didn’t try to chime in with my own thoughts on making the situation better, because I think it can be made substantially better than the book leaves you believing.

First off, we only need to get a little better to get substantively better decision-making. A shift of, on average, a few percent will move a repeated process like elections away from being a random walk and towards gradual improvement. From there it is a matter of the miracle of compound interest to carry us towards better and better outcomes. Congressional Representatives are elected anywhere from an average of five times to 25 times, for senior leadership. Senators are elected less often, but almost half of the Senate has served previously in the House. They have ample opportunity to learn new patterns of response to voter knowledge.

Smart Voters

We can start making voters smarter by expanding the franchise to smart voters. This means destroying efforts at voter suppression and gerrymandering. Voter suppression (such as onerous voter ID laws) are aimed at exactly the person we want voting – the motivated voter. Political Ignorance has demonstrated that motivation and knowledge move in proportion to each other, and argues for a causal connection.

Similarly, gerrymandering devalues votes, demotivating voters. We are moving towards mathematical definitions of gerrymandering that will help the courts decide when districts need to be redrawn.

Overturning the Citizens United Supreme Court decision is also critical. Part of the problem for voters today is that the signal-to-noise ratio in media is very low. Voters are inundated by low information fire hoses of advertising and fake news. The time has come to recognize not just financial institutions as “systemically important”, but elections as well – imposing stricter requirements than normal free speech.

Doing away with voter suppression, gerrymandering, and Citizens United will have a major impact in making voters better informed and bringing more motivated voters to the polls. But we can do more.

Better Candidates

Just being a candidate should put a burden of proof on individuals that they are fit for the job. In the past, this proof has been indirect, mainly being a member of a preferred class such as white, male, property owners of a certain age.

We can do better by mandating transparency in finances, physical, and mental health, depending on the risk and level of government service. For the President, the Personnel Reliability Program would be an apporiate test. This is the test that must be passed by nuclear launch officers. Considering that the president is the first link in the chain of launch authorization, (s)he should be able to pass the same test.

Audit the Fed

If we look at the history of American capital markets it is one of boom, crash, manipulation, and mistrust until the Great Crash and the formation of the SEC in response. Government regulation of markets and corporations brought stability to the capital market and greatly lowered the risk premium of investment, allowing corporations access to larger and larger amounts of capital. Contrary to libertarian claims, without government regulation, the capital markets would not have been able to function effectively and power the growth of the last century.

One key part of this success was the requirement for the publication of yearly and quarterly financial statements, audited by independent certified public accounting firms. These statements provide clear, detailed, and trustworthy information on the activities of the corporation. Investors rely on these reports and swiftly punish executives that manipulate them. Indeed, Donald Blankenship could have served more prison time for securities fraud than for killing 29 miners.

The learning point is that a similar source of trustworthy information on politicians and their accomplishments is not currently available or widely known. While financial statements for government units would be helpful, a wider, ‘triple-bottom-line‘ approach may have to be devised that would need to be applied quarterly to give voters enough data to work from.

Just as the SEC assign corporate executives CIK numbers to track them across corporate boards and influence, it would help to have numbers assigned to politicians to be better able to collect data on all of their activities. There can be no libertarian outcry here. We already do this for the sake of money. Doing it for the sake of liberty is an even stronger rationale.

Game the System

Political Ignorance refers to one study in which participants scored higher on a test of political knowledge when offered a cash inducement. Somin downplays the importance of this result, which I think is a mistake. The important point is not that it is impossible to pay people to vote. The important point of the study is that people have more knowledge than they use unless they can be motivated to use it. They don’t need to be motivated, and only then start to acquire knowledge. Some amount of unused knowledge is stored away waiting only for motivation to bring it forth.

Can voters be motivated to access this knowledge in the voting booth? I say yes, we just have to be creative in how we do it.

Let’s imagine a contest that any registered voter can enter. It starts with a test of knowledge in three areas, the structure of government, current issues in their district, and how to filter the media to recognize bias and fake news. The contest comes with additional apps, games, and referral bonuses to help a voter up their score. Take the test as often as you want, only your last score will count. To qualify, you must take the test within a week of voting.

The next part of the contest is a lottery that rewards participants randomly, but with bias by their score. Since there is no common way to demonstrate evidence that you voted in the US, the contest is not rewarding voting. It is rewarding being a well prepared potential voter.

By making the second part a lottery, we limit the cost of the contest while still motivating participants. This bounded cost could be undertaken by the Federal Election Commission, the League of Women Voters, George Soros, Facebook, the Koch Brothers or all of the above. Yes, there is a danger that partisan actors could put out their own contests with biased questions – is that any different than political advertising today? On the other hand, a billion dollar contest sponsored by Apple, Facebook, and Google with the LWV seal of approval has a powerful cachet.

Napoleon is supposed to have said, “A soldier will fight long and hard for a bit of colored ribbon.” We aren’t asking for death in battle, just voting knowledgably. Voters can be motivated, can be come knowledgeable, using the tricks and insights of modern marketing and psychology. We only need to make a small difference for the electoral process to become better and better. So as daunting as the task is, we have to try.

What is the proper size of government?

Ilya Somin’s book, Democracy and Political Ignorance, ends with a strong thesis statement. Political ignorance is a serious weakness of democracy, but we can reduce its dangers by limiting and decentralizing the role of government in society.

The government that governs least is not always best in every way. Yet it is the form of democracy least vulnerable to political ignorance. Democratic control of government works better when there is less government to control.

But we don’t have to imagine what America would be like with a smaller government. We can just look back at America throughout its history up until 1932.

The steamboat Helen McGregor, Capt. Tyson, on her way from New Orleans to Louisville, stopped at Memphis, on Wednesday morning, February 24, 1830. She had been lying at the wharf about thirty minutes, when one or more of her boilers exploded, with the usual destructive and melancholy effects. The loss of life by this accident was, at that time, unprecedented in the records of steam navigation. In the bustle incident to the landing and receiving of passengers, a part of the deck near the boilers was crowded with people, all of whom were either killed instantaneously, or more or less injured. No person in the cabins was hurt. The number of those who perished at the moment of the explosion is variously estimated at from thirty to sixty. As many of them were strangers whose homes were far distant, and whose bodies were never recovered from the water, into which they were projected, it is very plain that an accurate account of the number of the victims is not to be expected. (Lloyd’s Steamboat Disasters)

Small government means no safety requirements or inspections. Feedback devices to bleed off excess pressure (called governors or regulators, hint, hint) were still in the future.

The 19th century was dominated by the slavery debate. It paralyzed the Federal government and led to the projection of state power over other states through the Fugitive Slave Law, for example. We were unable to avoid the Civil War, massive loss of life and economic disruption.

Towards the end of the century, we saw the rise of monopolies, trusts, and company towns that strangled the free market and freedom to contract with coercive business practices. Patent medicines were peddled from town to town without concern for whether they worked the cures they were touted to produce.

Finally, the capital markets ran right off a cliff in 1929 under the influence of speculation in a new form of financial instrument – the mutual fund. The markets had always been lurching back and forth between runs on banks to crashes of the stock exchanges, but this Great Crash and subsequent Great Depression brought to a close the era of small government in America.

Promote the general welfare

The Preamble of the U.S. Constitution says that among the chief purposes of the Federal government are “provide for the common defense, promote the general welfare…” and these are specifically repeated in Article 1, Section 8. The quick review of one hundred years of history from 1830 to 1929 shows that the small government version of the Federal government active during that period did a spectacularly bad job of promoting the general welfare.

But what about the free market? Didn’t unfettered market capitalism solve the nation’s problems? No. Instead, the market was captive to various interests, including slavery. The capital market careened from ditch to ditch as speculators touted the high technology of the day – canals, railroads, telegraph lines – that ended in bankruptcy after bankruptcy. Far from protecting the country in the absence of a robust government, the markets could not even protect themselves.

Small and limited government is clearly a danger to the economy and nation it is meant to protect and promote if it is too small or too limited. So how do we ‘right-size’ government?

The proper size of government

If ‘small’ isn’t the answer, what is? How do we decide on the proper size of government?

One answer is that the proper size is zero. Various idealist philosophies and economic systems hold to a belief, unsupported by evidence, that human society doesn’t need government, and should move towards such a state.

While there is nothing to support such a belief in history, psychology, or group behavior let’s be charitable and just say that this is probably an ‘unstable equilibrium’. If by chance it was achieved, any perturbation would force society away from this state.

Another view might be that government should be a certain absolute size larger than zero, but not growing or changing as society grows and changes. For example, Congress is now fixed at 2 Senators per state and 435 Representatives. It doesn’t matter if the population grows or the economy grows, those numbers are fixed.

In the area of sports, the number of umpires or referees on the field is also fixed, but so is the number of competitors. For sports with complicated rules, the number of ‘regulators’ on the field can be 25% of the total participants.

While this might work for sports and legislators (and some think it doesn’t work) it obviously doesn’t work for meat inspectors, VA doctors, or weather scientists. Our government needs for these jobs varies with the economy, warfare, and the climate.

Most people accept that the size of government should somehow be proportional to the size of the nation. Here again, we can ask what is the constant of proportionality and what is the variable or variables we should use.

In Political Ignorance Somin usually refers to the ratio of government spending to GDP. He also at times refers to the complexity of government, noting the number of cabinet level departments and independent executive agencies.

The US currently runs a ratio of all levels of government spending to GDP of about 36%. During the response to the financial crisis, this was as high as 41%. For comparison, the ratio was about 7% in 1900 during the small government era.

Smaller economy, bigger government

Somin at times praises other, smaller countries for running capable democracies. He doesn’t present any data on political ignorance in these countries, however, so it is hard to see what the real argument is. Indeed, Luxembourg, Belgium, and Denmark all have higher government to GDP ratios than the US.


As the above chart shows, the US Federal government is already one of the smallest in the world relative to the size of the economy it controls. Switzerland is the only appreciably different government, and the difference between the US and Switzerland can be explained in large part by differences in military spending. Indeed, looking at this image it can reasonably be asked if the US government is still too small.

Exponential growth

As noted earlier, the ratio of all-levels government spending to GDP was 7% in 1900, compared to around 36% today. We might reasonably ask if the constant of proportionality is not, in fact, constant. Even taking account of war spending, is the ratio changing consistently over time, and if so, why? And is this a good thing?

One reason the ratio might be changing over time is that we’ve chosen the wrong variable. There are many variables that grow exponentially over time in a growing economy, instead of linearly. The idea of exponential value growth is commonly termed Metcalfe’s Law.

The implication is that as the economy grows, more and more of that growth takes place in the government sector. The total economy continues to grow, but the government is an ever increasing share.

For some people, this might bring to mind smothering and sclerotic bureaucracies that shuffle paper but achieve nothing. Certainly, that is a danger. But it isn’t inevitable.

Regulation of the Body (Not) Politic

As a counterexample, let’s take a look at the growth of regulation in genetics.

Some genes code for the direct production of a protein, the working chemicals of a cell. Since the beginnings of evolution, all cells use the same basic suite of proteins. It is this fact that allows us to trace the genetic relatedness of species.

When the human genome was first being mapped, the expectation was that it would code for up to 100,000 separate and distinct proteins and that these proteins would be what distinguished humans from other living things. It was therefore quite a shock to find that the human genome, as large as it is, only coded for about 25,000 proteins – the same basic repertoire as other animals.

We eventually found that what was different about humans was the gene regulatory networks of the human genome. Humans have hundreds of cell types as adults, and hundreds more exist as intermediate stages between the one cell type we all start as and the end product.

In each of these cell types, the same basic protein manufacturing is turned on or off in varying ways. Neurons build the machinery of synapses, stomach cells build other secretion functions. The big differences are not the proteins, but the regulations of the proteins.

This tremendous regulatory scaffolding allows humans to have large bodies of many different cell types. We couldn’t be who we are without the regulatory apparatus growing more important than the basic protein apparatus. Even the most basic one-celled animals, such as yeast, have ‘operons’ – genetic switches that regulate the production of proteins. The difference between us and yeast is a more sophisticated regulatory network, not some unique protein.

Going even further back into evolutionary time, bacteria have much simpler regulation of genetic copying than eukaryotes, such as plants and animals. The result is that their genetic code is copied with mistakes much more often. Bacteria can use this (“it isn’t a bug, it is a feature!”) to change rapidly in the face of environmental stress, such as anti-bacterial drugs. But when building a body of several trillion cells that has to last 100 years copying errors really are a problem.

Copy errors are corrected (or prevented) by systems that don’t create protein. These genetic regulatory systems are ‘takers’, not ‘makers’. If the human body was run on small government principles, we would still be bacteria.


It is also possible that instead of scaling on the number of things (GDP) or the number of connections (Metcalfe’s Law), the government scales on the number of types of things. Every innovation brings with it new cases and distinctions to be made, new kinds of behavior. A recent case had to answer the question of whether sending a text message urging suicide could be considered manslaughter.

Another aspect of novelty is the satiation of previous needs. If a government can help its citizens be well fed and avoid contagious disease, why not move on to avoiding pollution and obesity? We set new expectations of government. Entirely natural, but leading to growth in government nonetheless.

But no simpler

Albert Einstein is supposed to have said, “A theory should be as simple as possible, but no simpler.” I think this applies to government, specifically the size of government, a government being the embodiment of a political theory. Based on the examples given above, it is reasonable to expect that government will scale nonlinearly with some core aspect of society. This means that government may start small in a simpler time and less complicated society, but will grow – should grow – as that society grows and becomes more complicated. This is as close as we can get to answering in a blog post, “What is the proper size of government?”

10 Things Libertarians should unlearn

Recently, a libertarian acquaintance on Facebook linked to this Forbes article to support his positions. Now getting an article on the Forbes website is not as hard as getting published in the print magazine. There are lots of ‘contributors’ with tenuous qualifications, but this author is an economics professor. So herewith is a casual fisking of his ideas.

1 – The government cannot create wealth, jobs, or income. Wrong.

The government can create money. You’d think an economics professor would know that. “Priming the pump” was the old phrase. Further, the government can be the most efficient employer for some services. We certainly don’t want multiple private armies, and military forces are one of the largest expenditures of the government. And in the end, fighting a war does create jobs – for soldiers. The private sector can never create those jobs.
Is this eventually paid for with debt? Yes, but that debt is paid with inflated money. The actual interest rate on the debt is the stated rate minus the inflation rate. This can be negative at times, and other times less than the economic multipliers of putting the money in circulation.

2 – Income inequality does not affect the economy. Wrong.

That poor people spend all of their income is a sign they have too little income. Rich people can put money into investments, but they also put money into assets with inflated values that are not productive. A billion dollars invested in bonds goes eventually to work. A billion dollars spent on houses and superyachts has a much smaller economic impact.
More importantly, the income spent by a poor person is working today. Even a bond investment only really starts working economically once the bond’s intended use is finished. Therefore the net present value of the income spent by the poor person has a larger effect than the bond investment.
Other forms of investment have even more remote connections to the economy. Only stock bought at the IPO will provide cash to the corporation to expand. All other stock market activity just moves money between investors.

3 – Low wages are not corporate exploitation. Wrong.
The author here (and in other places) assumes that there is always another employer, always someone else ready to make a deal. This is fantasy free market thinking. It derives from the ideal world invented by economists making “simplifying assumptions” in order to use calculus. These assumptions include the ideas that everyone has perfect information, that markets are frictionless (no transaction costs), that there are an infinite number of buyers and sellers, and that everyone has the same desire to maximize net present value.

“Greedy businesses cannot exploit workers because another greedy business would be happy to exploit them a little less until greed removed all the exploitation.”

In the real world, McDonalds and Walmart can hold coercive monopoly power in communities, demanding low wages. There is no “other greedy business” in a small town where Walmart has put the Main Street businesses out of business.

4 – Environmental regulation is a regressive tax. Wrong.
The poor suffer more from pollution than other classes. That means they benefit more from cleaning it up. Calling any spending that is not scaled to income a “regressive tax” is BS argumentation.

5 – Education is not a public good. Wrong.
The lasting benefits of invention and innovation go far beyond what the inventor is paid. Perhaps the author is not aware that patent rights lapse after 17 years. The invention and all of its benefits do not evaporate when that happens. Educating people creates benefits far greater than their earnings.

6 – High CEO pay is no worse than high entertainer pay. Wrong.
CEOs are not entertainers. People voluntarily pay ticket prices to see an entertainer. Shareholders expect profits, and high CEO salary directly cuts profits, often dramatically. Through ESOPs, 401(k)s, pension funds, and other vehicles many workers are shareholders in the company they work for and in much of the American economy. They have a right to those profits. The author seems to live in a world where the investor class and the working class do not intersect. Maybe in 1875, not today.
Shareholders, whether workers at the company or not, might see that paying all workers a higher salary rather than one person would attract better workers and make better products. But CEOs and the corporate compensation committee are sold on the Napoleon theory of corporate leadership – the ‘great man’ theory of history applied to business.
7 – Consumer spending is not what drives the economy. Wrong.
Even the author has to admit that consumer spending is 70% of the economy. That might be higher than optimal, but then consumer saving is needed. It is still consumer behavior – individual choice.
Further, consumer behavior is the driver of innovation. This has been true since at least the invention of the automobile. Supercomputers that predict the weather use graphics processors and so do Bitcoin miners. Why? Because of video games. Drones, for better or worse, are enabled by the technology of smartphones. Delivering high-quality web maps and driving directions led Google to map every street in the world, a key enabler of self-driving cars.

8 – Government programs are low quality. Wrong
The first question to ask would be ‘compared to what?’ Private infrastructure frequently collapsed and the firms went bankrupt, that is how the government wound up owning many roads, bridges, railroads, ports, etc. The main job of government is public health. We have gone from epidemics of cholera and polio to epidemics of obesity and opioid abuse. That is high-quality government work!

9 – The government cannot correct cosmic injustice.
No, but it can minimize the cost. The same as any insurance pool works by spreading risk and profit, social spending works for all of us. If a very sick child keeps their parent out of work, out of college – that affects the entire economy through lost productivity. Insurance companies don’t correct cosmic injustice. That hurricane, tornado, or earthquake really did happen. They just help our entire economy recover more quickly from the loss.

10 – There is no such thing as a free lunch. Wrong.
A free lunch is a positive net present value. Many investments have a positive NPV, otherwise, we wouldn’t make them. Public health, education, and infrastructure are exactly those kinds of things. We saw earlier that the fantasy free market drives many of the author’s conclusions. In the fantasy free market, there can be not positive NPV because the cost of any investment already includes any possible profits the investor could make from it. This is implied by the perfect information assumption.

It is sadly typical of libertarians I interact with that they believe the market can do no wrong. Buildings don’t collapse in India because the market is perfect. Buildings collapse because the market is imperfect, humans can be greedy, and we need help.

Competition is great. But there aren’t only players out on the field, there are umpires and referees, also. In major league baseball, there is a maximum of 13 players on the field and four umpires. American football has 22 players and seven officials on the field. In basketball, there are ten players and three referees. Complicated games, many rules – if we want fair play there have to be regulators, even in a sport. Or we could just let the Yankees win.

Foot voting and political ignorance

Chapter 5 of Ilya Somin’s Democracy and Political Ignorance covers the interesting topic of foot voting and related issues. This is also a chapter where the author wanders significantly from the topic of political ignorance to discuss several aspects of the interplay of the structure of government and possible outcomes for the population. While many of these issues deserve discussion, they deserve more than the fragmented and passing coverage they often get here. Somin acknowledges that the book can’t address some of these topics in the depth they deserve, but it is still frustrating.

What is foot voting?

Foot voting is the heavy artillery of political expression.

To differentiate foot voting from any selection decision we have to focus on the actual change of location – the voter is physically moving to change jurisdictions.

To compare, a voter in the US might vote annually from 18 to 78 years old. There could be 10 decisions to make in each election, from choosing the President to ballot propositions. A dedicated voter might also vote in primaries. So a voter in the US might vote between 600 and 1,000 times during their lifetime. In contrast, adults in the US move only a few times during their lives. By one estimate, after 18, Americans move a little more than eight times, on average. (I think this estimate of number of moves is significantly higher than the one Somin uses in the book.)

Foot voters are exchanging all of government from some level down, even officials and policies not on the traditional ballot, with another package of officials and policies. If they meet residency requirements, they might even vote in the same year they move, getting even more power.

While the voter swaps governments, the governments gain and lose the voter. Not just the vote, but the economic contribution of that person and any associated family members.

Why foot voting?

Americans relocate for education, job opportunities, and to avoid negatives of the current locale. Historically, the size of these opportunities and negatives was larger than they are today. People moved to the frontier, blacks moved to the North to avoid lynching and murder – few of our moves today have such large motivational aspects. But to overcome the high transaction costs of moving, foot voters must be highly motivated, and that motivation level can translate into acquiring more political knowledge.

At this point, it would be great to actually have hard data to talk about. If surveys of public knowledge asked the questions “How long have you lived in your current residence?” “What are your current and previous zip codes?” we could actually compare levels of ignorance between recent arrivals and long-time residents. Somin does not present any data like this (it might not exist), so we are back in the world of making persuasive arguments rather than discussing data. For the same reason, we can’t examine whether foot voting is extra powerful, or less powerful, as the size of the move scales up or down.

Given that the evidence presented in Chapter 1 motivated the entire discussion to this point in the book, the lack of data to motivate arguments and conclusions about foot voting is strongly felt.

Advantages of foot voting

Once the ‘Bayesian prior’ of the decision to move has been established, Somin argues that several advantages compared to ballot-box voting are activated. First, the mover is acting as a ‘responsive voter’, so their information needs are low. The mover has access to ‘opinion leaders’ – their friends and family that may have already made a similar move and chosen a particular locale. They also can use information shortcuts such as party dominance in a locale to infer policy positions rather than researching each government official.

In chapter 2, we saw that even the responsive voter theory suffered from an excess of ignorance. In chapter 3, we saw that opinion leaders and information shortcuts were also inadequate to mitigate ignorance.  Somehow, somehow, the extra motivation of foot voting carries them over the line.

But does it really? These are good arguments, but without data that is all they are. There is also an unresolved tension between the idea that foot voters are highly motivated and therefore less ignorant, and the idea that whatever the level of ignorance foot voters actually have, it is irrelevant because the non-political benefits of foot voting outweigh the political considerations. The first of those is a testable hypothesis. The second is just hand waving in favor of a pet theory.

Retirement foot voting

In the current US economy, there is a special class of foot voting  – seniors making end-of-life retirement decisions. These decisions are made with a very different set of assumptions than other moving decisions. And given the growing size of the retirement population and their willingness to also ballot-box vote, this is a significant sub-population of foot voters.

I would expect that these differences skew results of some statistics looking at all foot voters as a class, for example their preference for small government, low tax and low service jurisdictions. If we subtract out this group the remaining foot voters would be looking for a substantially larger government, accepting a larger tax burden in return for higher services.

Neither group can be ignored, but we have to be careful about accepting an argument (Foot voters are well informed, foot voters chose small government, therefore small government is ‘obviously’ better) based on summary statistics that mask significant variation.

Private government

Speaking about retirement communities allows chapter 5 to segue to the topic of private substitutions for local government. There are many variations and gradations this substitution of private for public can take.

Sometimes there is talk of local governments allowing multiple service providers to operate within the jurisdiction, on the theory that they will compete. I once lived in a town with multiple garbage collection firms. Many places have Internet service through both phone companies and cable companies.

In practice, this does not usually work out as intended. Rather than compete, the firms form a loose cartel and divide up the business. In the town I used to live in, each garbage collector kept to their own streets, so each street effectively was a monopoly. In the town I currently live in it is impossible to get fiber optic Internet services from the phone company in apartment buildings that were built with an agreement to use cable internet. The result is still a single provider, with poor service and high prices.

This is no better than a government provider, and arguably worse from the point of view of transparency. Breaking up such a cartel is a high cost to residents, so it is usually tolerated. Chapter 5 does not discuss this point.

Some local governments have retained the responsibility to provide services, but outsourced the staff to do so. Sandy Springs, Georgia is an example. In this model, the argument is that the contract negotiation process allows the government to entertain competing bids and choose the lowest cost service provider.

In the case of Sandy Springs, it is too early to tell how this will turn out. The city founders have been diligent and proactive in contract negotiation. The city benefits greatly from being carved out of Fulton County, gaining a strong tax base but somehow avoiding responsibility for pre-existing debt and pension obligations. Nice work if you can get it, but not necessarily replicatible.

Sandy Springs seems to be getting FTE (full time equivalent) services for less that $100K per FTE, which is what I spitballed personnel costs would be in that part of Georgia. The outsourced employees are in places like Ireland and India.

Of course, the government can do the same thing. I deal with a Federal shared services center in West Virginia, a low cost locale compared to DC. Outsourcing is not the magic here, contract negotiation skills by the city management is.

Condos, developer communities and sprawling retirement venues such as The Villages are a different level of substitution. In these cases, the physical infrastructure is owned by a business entity. Residents can own units, but pay fees to the property management firm for basic services.

This is a complete flip-flop from the Sandy Springs model. In Sandy Springs the city was still owning the infrastructure, but contracting for competing service providers. In the condo or retirement community the infrastructure ownership is private and the service provider is a monopoly of the developer/property manager complex of firms. Competition happens at the level of the community vying with other communities for residents willing to relocate.

Implications of private government

Even huge developments such as The Villages are not governments. In a dispute with the IRS over bonds that were issued by The Villages as tax-free ‘municipal’ bonds, the IRS found that the board running the core community was merely an arm of the property owner – a private entity.

A citizen has certain rights that they retain, no matter what. A condo resident agrees to a business contract, often including the demand to use binding arbitration. To me, this is the biggest distinction between public and private – the substitution of citizen rights for contract stipulations.

On the ‘happy path’ of a contract, residents can focus on lower costs for services, if in fact they are lower. They might feel an emotional satisfaction of being ‘in control’ that comes from choosing and owning something, and this is no small thing. However, once the relationship sours the resident can find their freedom to obtain information or influence policy is sharply constrained.

This is important in a book about ignorance and the influence of the citizen. It feels as if a law professor could have done a better job of elucidating the differences between public and private rights. Chapter 5 doesn’t do this.

Rogue government

Sometimes governments do not serve their citizens well. Large corporations and sports franchises can play off communities against each other, seeking tax abatements and other forms of corporate welfare. This rent-seeking behavior works against voters’ interests. State and local governments can also raise exclusionary barriers with licensing and zoning.

Worse, we get communities such as Ferguson, Missouri. As was brought out in detail by the Michael Brown case, the government sustained itself on a tax-farming system of traffic tickets and bench warrants and maintained a sharp racial distinction between the governing officials and the citizenry.

Ferguson would seem to be a classic foot voting opportunity. The government was preying on its citizens. Well, Ferguson is a good example of the kind of foot voting known as white flight. Even as Ferguson’s population shrank across recent decades, the black population rose. The urban poor moved into Ferguson because it was even worse elsewhere.

Here we see the limits of foot voting. As pointed out earlier, the level of opportunity represented by the frontier, by the oppression of the Jim Crow South, doesn’t exist anymore. Arguments for foot voting based on trans-national flows motivated by famine and mass murder are irrelevant to what foot voting means in contemporary America. Instead, we have foot voting based on affordable housing with no particular consideration for the politics of the community. Foot voting doesn’t work if the prior is despair, not motivation. Good government has to work even for such citizens.


Chapter 5 is probably the weakest chapter in Political Ignorance. The author seems to have taken a well researched book about data, theory, and rationality, and inserted a topic that is at best marginally relevant. This topic is obviously the pet enthusiasm of the author, so it succeeds in ‘rent seeking’ more space with less justification than the other chapters. The author relaxes his standards of argument and throws out multiple justifications for foot voting, each with little support or evidence in its favor.

Democracy and Political Ignorance – Summary review

I finished reading Ilya Somin’s provocative book (henceforth Political Ignorance) over my summer vacation, and I think I want to write quite a bit about it so this will get spread across several blog posts.

Yes, readers concerned with the democratic process and constitutional design should read this book. It provides clear evidence that by whatever process you think democracy arrives at good governance of society, our current electorate is so woefully ignorant that it cannot be working properly. Further, this situation will be incredibly difficult to fix. While that might sound quite depressing, the author surveys many suggestions for trying and agrees that trying is necessary. Our basic government structure of federalism offers hope that experiments will have room to be tried, and successful experiments replicated.

No, readers should not be swayed in their decision by the subtitle, “Why smaller government is smarter”. Whether you agree with the sentiment or not, arguments about the ‘size’ of government make up a relatively small part of the book’s content. Political Ignorance does not make a serious attempt to address the question.

Somin presents the problem in chapter 1, with a review of recent surveys of voter knowledge from US federal elections, both Presidential and mid-term. Whether you look at all adults, registered, or likely voters, the results are appalling. Easily 25% of the country just has no fucking clue, just no clue at all about headline issues facing the country that their vote might affect. A key finding, repeated later in chapter 4, is that swing voters are the most ignorant of all.

Let that sink in – our elections are being decided by the dumbest of the dumb.

If I had any wishes for chapter 1, I would wish for a longer baseline of data and breakdown between factual knowledge of the issues vs. a knowledge of the system.

Chapter 2 examines four different theories of democratic governance, each of which seems to hold voters to a different standard of knowledge about the issues. “Responsive government” assumes democracy is a ‘throw the bums out’, reactive process. Old leaders are punished for large failures such as losing wars or causing famines, and replaced with new leaders. This is certainly a step up from monarchy or other forms of dictatorship, but suffers from many emotional biases and information deficits. What leader or policy was responsible for the problem? Can Chris Christie be blamed for shark attacks driving down tourism on the Jersey Shore? (Answer: yes.)

Burkean trusteeship holds that voters are actually selecting trustees for the work of governing society, and as such the important part of the process is choosing the person for an elected office according to their qualities of education, personality, moral compass, etc. Here again, we find that voters know very little about the people they are electing. Think birtherism for an example.

Issue-driven majoritarianism simply asks if a candidate is siding with the majority on specific issues. In this view, reflecting the majority desire is more important for democracy than any attempt to lead the entire nation. However, the likelihood that a poll-following politician will be elected is confounded by the ignorance of voters on the issues, who holds what position, and who is relatively closer to the majority view.

Finally, deliberative democracy assumes that voters are researching and discussing the facts, the proposed policies, and the likely outcomes in an informed and conscientious manner before stepping into the voting booth, and exercising their franchise with the solemnity deserving of the act. Obviously, this is not happening. Exchanging Facebook memes with your friends does not count as deliberative democracy.

While this survey of process theories is not exhaustive, it does show the span of levels of knowledge that different theories can assume. Even the most lenient, reactive model won’t work if voters don’t know who to blame for what. However, we can take some small comfort from work on the evolution of cooperation, such as Axelrod. In this research we can see that the best strategy for the Iterated Prisoners Dilemma is Tit-for-tat, the reactive model. If we view government as a long IPD held between the class of politicians that run for office and the class of voters that elect them, then there is small hope that over long periods of time the system can evolve to stable payoffs. Is this ‘perfect’? No. Is it better than revolution? Yes.

Hello Darkness, my old friend

In chapter 3, Somin introduces the idea of rational ignorance. Since everyone is working with scarce resources of money, time, and energy, it is necessary to choose where to invest those resources. For most voters, becoming politically knowledgeable is not the best use of their resources.

If we compare voting to other selection processes such as jury voting, product buying, and choosing where to live it is easy to see a proportional relationship between personal motivation and acquiring knowledge. If we value the outcome of the decision more, we will be more likely to devote the resources necessary to acquiring the knowledge to better make the decision.

From this perspective we can see that modern American voters just don’t think the act of voting will change very much in their lives. Voters that are choosing freedom or slavery, monarchy or communism, have a lot at stake, and consequently political discussion dominates the public square in a more informed manner than when thinking about Medicare Part D.

If this ignorance is rational, Americans have a pretty good thing going and they doubt that any one election can change that very much. Most of the time, they are right. But there is a ‘long tail’ of times they are wrong, and being wrong in those decisions means being catastrophically wrong. The price of freedom is eternal resource allocation at higher than rational levels.

What can be done about this situation? Americans are so fat and happy, the system has worked so well for the majority, that we have fallen into a post-Thanksgiving-dinner coma. If only we cared as much about politics as we do about college football, we might wake up!

Chapter 4 looks at some workarounds to ignorance. Maybe we are smarter than we seem at first glance, or have the answers hidden up our sleeve. If we can be more informed for a lower cost, that would be a good thing. Some of the strategies that are looked at include political parties, opinion leaders, knowledge from ordinary life and retrospective voting. Sadly, none of these are theoretical possibilities that have not been tried. The existing data showing extreme political ignorance already has all of these strategies and more already baked into the result. Shortcuts will not help us.

I’m going to save Chapter 5 for a separate post. Foot voting is obviously a pet enthusiasm of the author’s. Chapter 6 looks at the concept of judicial review with respect to political ignorance. This has the feel of a separate essay, and can safely be skipped if legal theory is not your thing.

Chapter 7 asks if voters can be made smarter somehow. Again a range of possibilities are examined, including shrinking the franchise to the well-informed through tests. This chapter can help dispel the depression brought on by constant drumbeat of negative results from the previous chapters. There is hope, because these and other strategies are just starting to be tried.

Political Ignorance has two punchlines. We can reduce the dangers of political ignorance by limiting and decentralizing the role of government in society, and democratic control of government works best when there is less government to control. I can understand why Professor Somin has buried these thesis statements at the end of the book, rather than presenting them at the beginning. This book, as valuable as it is, does very little to prove either statement. Again, more on this in another post.

Bottom line – this book is an important survey of a difficult topic. It will reward your scarce resource allocation! Buy it, read it, talk about it.

Democracy and Political Ignorance

I’m in the process of reading a fascinating book on the political process, and the lack of rationality in democracies. The book is Democracy and Political Ignorance: Why Smaller Government Is Smarter, by Ilya Somin. While the American experiment in democratic government has always recognized the dangers of mob rule and populist manipulation, our understanding of these issues has sharpened in recent years as political scientists dissect the issues of education, influence, class, gender, and race on voting – the core act of creating a government.

For many Americans, there is no more apt time than now to be thinking about these issues. Well, perhaps two years ago would have been more apt.

I can tell you in advance that I find the subtitle provocative. I don’t ‘believe’ in small government or big government, I want a right-sized government for its expected function. “Smaller” government can be like “thinner” for an anorexic – you can always be thinner.

Nor is smaller government fully compatible with limited or decentralized government, other qualities that the book (and its author) advocates. The experience of the capital market is that there really is such a thing as ‘economies of scale.’ If you want make organizations smaller for the same set of functions, merge them and fire the middle management, that is the lesson of corporate raiders, outsourcers, and shared service centers.

It’s great to read a well-written book on an important topic, and even better if you don’t agree with the author 100%. I’m looking forward to sharing more of those disagreements here.