Political Time Horizons

A common argument in favor of government actions is that they are necessary to make us take account of long run consequences. As best I can tell, this is precisely backwards.

To see why, consider why someone might invest in planting black walnut trees, which produce valuable lumber but take a long time to mature. A sixty-year old investor might plant them even if they take forty years to mature because he can sell land with twenty-year-old black walnut trees to somebody willing to wait another twenty years before harvesting them. That is why, in ordinary private markets, rational people find it in their interest to make long term investments, including ones whose payoff will take longer than they expect to live.

This assumes secure property rights. Suppose there is a significant chance that the government will decide that black walnut trees are a national resource and expropriate them or that some private entrepreneur will decide that black walnuts are more valuable in his hands than in mine and come in the dead of night with a chainsaw and a truck and drive off with my trees, not yet fully mature but already valuable. The higher that probability is, the greater the payoff has to be before it pays me to plant.

Politicians have insecure property rights in their political assets, a reasons for them to heavily discount long term consequences of their decisions. If Biden does something politically costly that produces desirable results twenty years from now he will probably not be the one who will get the credit. One way of interpreting Obama's repeated claim that anyone who liked his insurance could keep it, a claim he surely knew the voters would eventually discover was false, is that he steeply discounted a political cost that would only come due after his final election.

A politician can claim that costs of his policies are the price for future benefits — but he can do so whether or not it is true. Knowing whether claims about the distant future are true is hard. A voter, knowing that his vote is unlikely to change the outcome of an election, has little incentive to be well informed in his voting; it is rational not to acquire information that costs you more than it is worth. Josef Stalin could not plausibly tell the people he ruled that they were living well but he could and did tell them that their current hardship was the price of catching up with and surpassing the capitalist West. It worked, for a while.

Consider four examples of short time horizons for political actors:

Pensions or Pay

You are the governor of a state threatened with a strike by  public employees, teachers or police. To raise their wages you would have to increase taxes or cut some other expenditure, both likely to be unpopular with your voters. If instead you agree to raise their pensions the cost will come in the future, most of it when someone else is governor. California has about eighty thousand retired government employees whose pensions are more than a hundred thousand dollars a year.

This approach to pushing costs into the future may be limited by requirements in state law that pensions be funded well enough to cover their future obligations. The future is, however, uncertain. A pension fund that does its calculation with optimistic assumptions about what its investments will yield — difficult with low risk investments such as government bonds, easier with riskier investments such as stocks or real estate — can hold down the cost in the current budget. When the investments turn out to do less well than predicted the difference between what the fund owes and what it has will have to be paid by the taxpayers.

Current estimates of unfunded state pension obligations range from a little less than a trillion dollars to over four trillion. More than two-thirds of the assets they manage are “allocated to risky investments like equities, and alternative vehicles, including private equity, real estate, and hedge funds.”¹

Social Security

Here is a simplified picture of how social security works — to avoid complications, I assume an interest rate of zero.

The system collects taxes and pays benefits. At any instant the system has money that has come in as taxes and not yet gone out as benefits. By gradually spending down that asset the system can pay each individual more than he put in — but when it has been run down to zero payments have to be cut or taxes raised. That can be put off for a while by expanding the pool,² making more and more people liable for the tax and using the money that has been paid in and not yet paid out to pad the pensions of those already in, but eventually everyone will be in the pool.

The projected reserve depletion date for the combined OASDI trust funds is 2034 … . … After the projected trust fund reserve depletion in 2034, continuing income would be sufficient to pay 80 percent of program cost, declining to 74 percent for 2097.

(The annual report of the Social Security Board of Trustees)

At any time in the eighty-two years since Social Security was created, payments could have been reduced to a level that could be sustained forever, a smaller reduction the earlier it was done. It didn’t happen.

The Debt Limit

Politicians like to spend money, usually popular, without raising taxes, usually unpopular. Borrowing is a way to do so. The cost will eventually have to be born but by that time someone else will be in office.

A debt limit, currently in the news, is an attempt to deal with that problem. In theory it lets politicians at time A constrain politicians, including themselves, at time B. It doesn’t work because when time B comes around there are always political reasons why the limit has to be raised.³ Some of the reasons are good politics but bad arguments. One example is the claim that if we hit the debt limit the government will be unable to pay Social Security.

When the Social Security system runs a surplus, as it did from 1984 through 2009, the money is lent by the Trust Fund to the Treasury in exchange for Social Security Trust Fund special bonds, which count as part of the national debt. Consider the following simple scenario:

1. The government hits the debt ceiling and can no longer borrow.

2. The Social Security Trust Fund asks the Treasury for ten billion dollars to pay Social Security recipients.

3. The Treasury takes ten billion dollars of revenue which it was planning to spend for something else, such as salaries for government employees, and gives it to the Trust Fund, redeeming ten billion dollars of the bonds representing its debt to the Trust Fund.

4. The national debt is now ten billion dollars lower, so the Treasury can borrow ten billion dollars and use it for the salaries it was planning to spend the first ten billion on.

This procedure is workable until all the bonds are redeemed. Since the Trust Fund is currently over 2.8 trillion dollars, that is going to take a while.

For more details, see this 2011 article by Thomas Saving, who served two terms as a trustee of the Social Security and Medicare Trust Funds.

The Debt Limit and Default

When there is controversy over raising the debt limit, many people talk as though if it is not done the government must default on its debt, stop paying interest. That is not true — default is neither necessary nor sufficient. It is not necessary because federal revenue is more than seven times net interest payments, more than enough to continue paying interest on the debt — provided the revenue isn't all spent for other things. It is not sufficient because net interest payments are about 640 billion dollars a year, the current deficit about 1,400 billion a year, so if the government stopped paying interest it would still have to cut other expenditures by more than 700 billion dollars a year .

What would it take to balance the budget? Last year’s federal expenditure was $6.3 trillion, this year’s currently projected at $6.2 trillion, so if the government continued to pay interest on the debt and reduced every other expenditure by about 23%, the budget would balance. I do not expect it to happen, but simply as a matter of logical possibility the government could balance the budget while continuing to pay interest on the debt and cannot balance the budget by defaulting on that interest unless it also sharply reduces other expenditure.

Iran: kicking the Can Down the Road

Doing something substantial to keep Iran from developing an atomic bomb,⁴ possibly risking a war, would have been politically costly. It made more sense to do smaller things, with luck slow the process a little, in the hope that that by the time they got a bomb someone else would be in the White House. When someone else was in the White House, the same logic applied.⁵

1. Pew research, reported in a CNN story.

2. Details here.

3. It might work if the limit took the form of a constitutional amendment, which is much harder to change, but so far that has never happened.

4. This assumes that the Iranians are trying to develop nuclear weapons. One reason to believe they are is that having a nuclear weapon is pretty obviously in their interest. Another is that Iran has large supplies of both petroleum and natural gas, hence no good reason to spend a lot of money and antagonize most of the developed world by building reactors, which is what they claim they are doing.

5. My own view is that U.S. foreign policy throughout my lifetime has been too aggressive rather than not aggressive enough but the stated view of my government was and is that Iran had to be kept from getting nuclear weapons. Doing that would require immediate political costs in exchange for a long-term benefit, which is why it didn’t happen.