Strange Economics
“God, grant me the serenity to accept the things I cannot change, Courage to change the things I can, And wisdom to know the difference.”
This is the famous serenity prayer, authored by Reinhold Niebuhr. It was the closest to theological orthodoxy that Niebuhr ever came.
It applies to voting.
VOTING AND LOTTERIES
Voting is irrational, according to free market economic theory. It is like buying a lottery ticket, only dumber. There is a Wikipedia entry on this: the paradox of voting.
The paradox of voting, also called Downs paradox, is that for a rational, self-interested voter, the costs of voting will normally exceed the expected benefits. Because the chance of exercising the pivotal vote (i.e., in an otherwise tied election) is minuscule compared to any realistic estimate of the private individual benefits of the different possible outcomes, the expected benefits of voting are less than the costs. The fact that people do vote is a problem for public choice theory. It was first observed in the modern social choice literature by Anthony Downs.
Was Downs correct? Economists think so. They generally dismiss buying a lottery ticket as a loss-producing activity for people who are bad at math.
Physical Gold & Silver in your IRA. Get the Facts.
A lottery is a form of gambling. Gambling is a zero-sum game: every winner wins at the expense of the losers. Also, there are house rules. The games are set up so that the house usually wins. This is why states run lotteries but prohibit private gambling: “the numbers.” The states want to steal from the poor and the ignorant. They resent competition from profit-seeking businessmen: bookies.
The person who buys a lottery ticket believes that he may profit greatly if his “number comes up.” Throughout history, men have believed in four incompatible systems of overall causation: fate, luck, randomness, and the sovereignty of God. Fate is seen as impersonal. Luck is seen as personal. Randomness is seen as impersonal. God is seen as personal.
People who gamble trust in luck — “lady” luck. Those who believe in randomness do not gamble. They understand the math. Those who believe in the sovereignty of God do not gamble. They trust in hard work, ethics, and the grace of God, none of which involves gambling. As for those who believe in fate, causation is meaningless anyway. They think men cannot affect the outcome.
The economist tells the gambler that the probability of a favorable outcome from buying a lottery ticket is so low that it is not statistically different from not buying the lottery ticket at all. “Save your money. Don’t buy the ticket.” But poor people buy them — the very people who cannot afford the loss. They have given up hope in ethics, hard work, or the grace of God. But they want to believe in something that may help them. They believe in luck. Buying a lottery ticket is a religious act for them: an objective way for them to register their faith in luck. Luck may “smile” on them, they believe. “Give luck a chance.”
I don’t gamble in zero-sum games. That is because I do not believe in fate, luck, or randomness. I am a Calvinist.
I did participate in a zero-sum non-game: the U.S. government’s allocation of radio spectrum that it stole back in 1927. I became a millionaire. So did my wife. We won two separate lotteries. My friend Terry Easton had figured out a way to beat the lottery. My subscribers won, and they won big. But my wife and I won biggest of all. It cost us nothing to play. My partner John Mauldin and his wife also won with us.
The government should have auctioned off the spectrum: “high bid wins.” But since it selected a lottery, I participated. But this was not a game.
The same logic applies to commodity futures. It looks like a zero-sum game, but it isn’t. The futures market solves two specific non-game problems: (1) how to allocate the ownership of uncertainty — speculators vs. hedgers — and (2) how to gain a better knowledge of future prices at no expense to society — “free riders.”
My point: sometimes something that looks like gambling really isn’t. Is this also true of voting? I think it is. But first, I will consider the costs of voting.
THE COSTS OF VOTING
Voting takes time. The more valuable your time, the more expensive voting is for you. This is the crucial economic insight regarding costs: the economic value of your highest-value use of a scarce resource. What would you have done with the time it takes you to vote? That is the cost of voting for you.
Libertarians have a mnemonic device: TANSTAAFL. “There ain’t no such thing as a free lunch.” But what about TANSTAAFE? “There ain’t no such think as a free election.”
Voting takes automobile depreciation: 56 cents a mile, the U.S. government says. It allows this for business depreciation.
It takes the time to study the issues. People watch candidates debate. Why? Their vote will not affect the outcome.
The news media are filled with information about the candidates. This displaces news from which readers and viewers might actually be able to benefit.
The outcome of your vote is statistically irrelevant. The cost of your voting is not statistically irrelevant. This is the strange economics of voting.
“YOUR VOTE DOESN’T COUNT.”
Critics of individual voters argue that voting is statistically comparable to buying a lottery ticket but dumber. The person who buys a lottery ticket has a statistical change of getting an immense payoff. A person whose party wins an election knows that his vote will not make a difference, unlike a lottery ticket. His party will win or lose, with his vote or without it.
Next, the payoff is far smaller than winning the lottery. He is way down on the food chain of victory. Politics is a popular form of trickle-down economics. The people at the top get most of the takings. Politics is mainly about divvying up stolen goods.
Statistically, your vote doesn’t count except in very small local elections. The larger the number of voters, the less your vote counts.
Voters in the USA care only once every four years. They care only about who is elected President. They vote in large numbers. So, the more important the election is to most voters, the less their vote counts. This is the strange logic of voting.
Their vote counts most in special elections held for a local office or ballot measure. But hardly anyone participates in special elections. That is why their vote would count far more. But they do not care. This is the strange logic of voting.
The post Strange Economics appeared first on LewRockwell.
Leave a Reply