The Becker-Posner Blog

August 2, 2009

Health Reform and Obesity--Posner

The biggest problem besetting the Administration's program of health reform is how to pay for it. The heart of the program is extending insurance coverage to tens of millions of people who at present are not insured. This will cost more than $100 billion a year just in subsidies, but the total cost will be higher because demand for medical services will rise. At present, people who are not insured are billed directly for medical services. Often they cannot pay, but then their credit takes a hit, or they are forced into bankruptcy. And emergency rooms use queuing to increase the cost of their services to the indigent. When the uninsured become insured, the marginal cost of medical services to them falls to the copayment or deductible that they are charged; the total price (pecuniary plus nonpecuniary) is now much lower, so more service is demanded, and prices to all consumers of medical services rise because supply is inelastic.

Some advocates of extending coverage argue that it will reduce aggregate medical costs. They point out that people may defer preventive care that might ward off an illness, or a worsening condition, that might cost more to treat than preventive care would have cost. The other side of this coin is that preventive care may keep alive people who would have died, thus ending their demand for medical care. But everyone dies eventually, and a very high fraction of total medical costs are incurred in the last few months of life. Moreover, because of technological progress and the high value that people place on extending their life, medical expenses are growing far more rapidly than per capita income, and, as a result, postponing death imposes disproportionately greater costs on the next generation. A partial offset, however, may be that greater and therefore more costly efforts may be undertaken to postpone death the younger the dying person is.

Preventive care can also be very costly, especially when it takes the form of expensive screening: screening costs are incurred by the healthy as well as the sick.

The most attractive form of preventive care, at least from a government budgetary standpoint (disregarding for a moment nonpecuniary benefits and costs, to which I'll return), is behavioral change: for example, safe sex as an AIDS preventive--or losing weight, or, more realistically, not gaining excessive weight in the first place, to prevent obesity.

Obesity has increased rapidly in the United States, to the point where, at present, more than half the adult population is overweight and 25 percent is obese. A recent study estimates that the average obese person incurs annual medical expenses that exceed by 42 percent the average annual medical expenses of the non-obese; the aggregate excess cost is almost $150 billion a year. Average expense is potentially misleading because of the shorter lifespan of unhealthy people. However, I believe that except in cases of extreme obesity, the effect on lifespan is less than the effect in creating medically treatable conditions such as diabetes, joint problems, complications from surgery, and cardiovascular disease.

The economist Tomas Philipson and I have written about the economics of obesity. We have pointed out that the decline in the price of fatty foods, along with the rise in the opportunity cost of physical activity (work is more sedentary than it used to be, so one has to invest extra time to get exercise, and television and video games have increased the utility that people derive from sedentary leisure pursuits), explains the dramatic long-term increase in the percentage of Americans who are seriously overweight.

It might seem that if people derive greater utility from consuming fatty foods in large quantity than the costs in illness and medical care, the increase in obesity actually is optimal from an economic standpoint. But there are three reasons to doubt this. The first is that the obese externalize part and probably most of the excess medical costs that their condition imposes, because health insurers (including Medicare) generally do not discriminate on the basis of weight. The second reason to doubt that we have the optimal amount of obesity is that high and rising aggregate health costs, because financed to a large extent by government, are contributing to the serious fiscal problems of the United States: the United States has a soaring national debt that may have very grave long-term consequences for America's prosperity. Obesity thus has potential macroeconomic significance.

Third, there is reason to doubt that the obese actually gain more utility from the behaviors that contribute to their obesity than the costs of obesity, which are not limited to medical costs but include discomfort, loss of mobility, discrimination by employers, and social ostracism by people who consider obesity repulsive or believe it signals lack of self-control, gluttony, or low IQ (or all three characteristics).

Obesity is highly correlated with education. Highly educated people are much more likely to be thin than people who are not highly educated. This is partly but not only because highly educated people have on average higher incomes than other people. They can afford more expensive foods, which are low in calories, and the cost of exercise, which can be considerable, as it may require joining a gym or having a personal trainer.

But income is not a complete explanation, because highly educated people in low-paying jobs, as many teaching (including college teaching) jobs are, tend to be thin. But is this because one needs education to realize that eating fatty foods makes one fat and that fat people have medical and other problems that thin people do not? Surely not. It is rather that educated people have better impulse control, or, in economic terms, a lower discount rate (the rate at which a future cost or benefit is equated to a present cost or benefit), than uneducated people do, on average at any rate. To get an education means incurring present costs for future benefits, and that is less attractive the higher one's discount rate. Moreover, intelligent people derive greater benefits from education in terms of present enjoyment and future income than unintelligent people do, and intelligence implies lower costs of foreseeing consequences of one's actions: it is easier for an intelligent person to realize the consequences of indulging one's tastes for fatty foods than an unintelligent person, given that obesity is not an immediate consequence of eating such foods. Low-IQ people (and many high-IQ ones as well) may also fail to realize how much more difficult it is to lose weight than to avoid gaining weight in the first place.

A further problem with people of low intelligence and (what goes with it) low income is poor parenting, as a result of which children grow up with bad eating habits, including excessive consumption of fatty goods; these habits may be difficult to break in adulthood.

If the unintelligent experience greater costs of imagining the consequences of eating fatty foods, that is an argument for providing them with greater information about those consequences, to offset their deficit in understanding. Maybe with full knowledge the unintelligent would be willing to incur the costs, in somewhat more expensive food and in fewer sedentary leisure pursuits, of avoiding becoming obese. So aggregate utility might actually be increased, as well as aggregate medical costs reduced, by an effective campaign of warning people about the consequences of eating fatty foods. I do not think that government should regulate behavior on the premise that it knows better what makes people happy than people themselves do; but controlling external costs is or should be an uncontroversial governmental function.

Such an educational campaign as I have suggested would be a cheap form of preventive care, but would it be effective? The evidence is mixed, but a 2008 review article by Lisa Harnack and Simone French in the International Journal of Behavioral Nutrition and Physical Activity finds that labeling restaurant menus with calorie information does reduce consumption of high-calorie foods. Conjoined with reduced calories in school lunches, elementary- and high-school courses in nutrition, and warnings in food advertising and labeling similar to the warnings in cigarette advertising and labeling, the prevalence of obesity might be reduced at slight cost--possibly to the benefit of almost everyone except the sellers of fatty foods.

One of the health-care-reform bills pending in the Senate would relax legal limitations on "discrimination" by private group-health insurers; that is a step in the right direction, as are growing efforts by employers to encourage their workers to control weight (the motive is to reduce the cost of health insurance to the employer). Medicare could be modified to reduce fees to thin people. In addition, a calorie-based food tax (which would, for example, fall heavily on sugar-flavored soft drinks), would reduce obesity at negative cost to the public fisc. Such a tax may seem "unfair" to people who consume such foods but are thin, but this is just to say that the tax would be at once a regulatory and a revenue tax, and in the latter aspect would be subject to criticism only if it were an inefficient tax relative to alternative methods of taxation.

Posted by Richard Posner at 8:00 PM | Comments (16) | TrackBack (0)

The Growth in Obesity-Becker
The weight of the average male and female began to grow sharply around 1980 not only in the United States but also in all other developed countries. When average weight grows, the rate of obesity-usually defined by a biomass index (BMI) of over 30- grows at a much faster rate. As Posner indicates, two of the important causes of the increase in weight and obesity are the rapid decline in the price of fast and fatty foods that began about 30 years ago, and the growth in sedentary activities, mainly driven by the expansion of television, and the development and spread of personal computers and the Internet.
Posner believes that consumer ignorance of the health consequences of their eating and sedentary activities also contributed significantly-particularly since lower educated persons have the highest incidence of obesity. I am doubtful, however, if ignorance of these effects has been important. Poor information is a last resort crutch that economists are increasingly relying on to explain consumer behavior that they fail to explain in other ways. For example, "behavioral economists" are arguing that many consumers run up large credit card debts in good part because these consumers are not aware of the level of interest payments, and that poorer borrowers took out mortgages during the housing boom years of a few years ago because they could not calculate the difficulty of meeting monthly payments. In both these cases, virtually no evidence is presented to support this thesis. In the eating case, most reasonably well-executed studies find quite small effects on eating patterns of providing nutritional and other information about foods.
Continuing with an emphasis on the relation between obesity and low education, Posner argues that heavy discounting of future consequences induced many less educated teenagers and adults to overeat relative to the adverse consequences on their future health once cheap fast foods became available. Perhaps insufficient attention to future health consequences has been important; although many other possible explanations are available for why more educated persons eat smaller quantities of fast foods and get more exercise.
Another change, however, also emerged around 30 years ago that provides a fully rational forward-looking incentive to pay less attention to the future health consequences of overeating and weight gain. I am referring to the beginning of the age of blockbuster drugs that help control blood pressure, cholesterol, and erectile dysfunction, help treat if not cure various cancers, and provide other protections against some serious health consequences of being overweight. The expectation of even further progress in the future, such as in treating the worst aspects of diabetes, would rationally reduce present concern about weight gain and the future consequences of heavy eating of rich foods and low levels of exercise.
This argument does not presuppose that persons with less education and even the more educated are keenly aware of these developments in drug therapy, or that they could articulate this as a reason for their eating patterns. All that is required is that most people have a loose awareness of the growth of drug therapy for many diseases, and that such awareness helps relax their concerns about gaining weight. To me that is a reasonable presumption that would producen rationally heavy "discounting" of the future health consequences of becoming obese.
Another argument made for public policy to discourage obesity is that obese individuals make demands on the health care system that raise the cost of this system to others. This argument is not persuasive to the extent that it relates to private health insurance. If obesity makes as large demands on medical care as claimed by the Centers for Disease Control and Prevention study cited by Posner, it would be in the self interest of insurance companies to charge significantly higher premiums to overweight persons since that would better align their revenues to the cost of treating obese persons. Moreover, it is easy to use weight as a determinant of insurance premiums since weight is easily measured in any physical exam required to get insurance. That private insurance companies do not use weight as a premium determinant means either that they do not consider the effects of obesity on their cost to be that large, or that they are afraid they will be accused of discrimination if they do use weight as a criterion. Perhaps the health bill in Congress would make it easier for insurance companies to punish overweight persons relative to thin ones.
Since taxpayers finance Medicare and Medicaid, these organizations do not have the same incentives as private insurance companies to penalize overweight persons for their excess weight. The cost imposed by overweight persons on public expenditures has been one of the justifications for taxes on fast foods and soft drinks since these are important inputs into weight gain. However, such a tax would be inefficient, perhaps highly inefficient, because it targets all persons who eat fast foods and drink sodas, yet most of these persons do not become obese. It is akin to taxing the sale of wines and liquor to reduce drunk driving, even though most drinkers do not drive drunk and cause accidents. This distinction between taxing inputs into drunk driving and taxing drunk driving explains the tendency to heavily punish people who are drunk, especially when they cause accidents.
The corresponding approach with regard to weight would be for Medicare to institute surcharges for very overweight persons (or discounts for thinner persons) and for Medicaid to impose various costs on obese persons who use their services, such as requiring them to spend time at educational classes on the control of weight, or to pay a fee for any Medicaid services they receive. Such charges and fees probably would run into strong political opposition, but they point the way to more appropriate ways to discourage obesity that causes medical problems that utilize public funds.

Posted by Gary Becker at 7:32 PM | Comments (13) | TrackBack (0)

July 26, 2009

Mortality from Disease and the American Health Care System-Becker

Many Democratic Congressmen, member of the Obama administration, and others writing about American health care envy the 'European" health deliver system since they attribute Europe's lower mortality rates, despite much lower per capita spending on health care than by the US, partly to the European model of health care delivery. This model involves government domination of spending on medical care that involves extensive government regulation and rationing of access to medical care. Envy of the European model explains why the Democratic proposed "reforms" of the American system involve large increases in government involvement in health care, including a government-run health insurance plan.
In evaluating whether envy of the European approach is justified, it is crucial to determine whether the higher mortality rates in the US than in many European countries is due to defects in the American health delivery system, or to other factors. Mortality rates are affected not only by health care, for they are also very much dependent on personal behavior, such as smoking, eating habits, exercise, stress, how carefully individuals follow the medical advise they receive, and many other kinds of behavior under the control of individuals rather than the medical profession. The US has relatively high incidences of obesity, partly because Americans consume lots of high fat and high cholesterol foods, and Americans were heavy smokers in the past, just to mention a few unhealthy forms of behavior. Perhaps then the higher US mortality rates are due much more to differences in personal habits and personal care than to defects in the US health delivery system?
One way to separate health care from personal behavior is to consider survival from serious diseases, such as various cancers and cardiovascular diseases. In my post on health care on June 7 of this year I referred to a study published in Lancet in 2007 that compares five-year cancer survival rates for the US, the United Kingdom, and the European Union as a whole. The study examines early diagnosis, early treatment, and access to the best drugs, and finds that the United States does very well on all three criteria. As a result, five-year cancer survival rates are much better in the US: they are about 65% for both men and women, whereas they are much lower in these other countries, especially for men.
Early diagnosis helps survival, but it may also distort comparisons of five or even ten-year survival rates since some cancers would be discovered at very early stages. An alternative that avoids this distortion is to compare age-adjusted mortality rates for different diseases. Early detection and other medical care that improved life prospects would show up as lower mortality rates. A recent excellent unpublished study by Samuel Preston and Jessica Ho of the University of Pennsylvania compare mortality rates for breast and prostate cancer. These are two of the most common and deadly forms of cancer-in the United States prostate cancer is the second leading cause of male cancer deaths, and breast cancer is the leading cause of female cancer deaths. These forms of cancer also appear to be less sensitive to known attributes of diet and other kinds of non-medical behavior than are lung cancer and many other cancers.
These authors show that the fraction of men receiving a PSA test, which is a test developed about 25 years ago to detect the presence of prostate cancer, is far higher in the US than in Sweden, France, and other countries that are usually said to have better health delivery systems. Similarly, the fraction of women receiving a mammogram, a test developed about 30 years ago to detect breast cancer, is also much higher in the US. The US also more aggressively treats both these (and other) cancers with surgery, radiation, and chemotherapy than do other countries.
Preston and Hu show that this more aggressive detection and treatment were apparently effective in producing a better bottom line since death rates from breast and prostate cancer declined during the past 20 by much more in the US than in 15 comparison countries of Europe and Japan. US death rate rates from prostate cancer went from about 7% above those of the comparison countries in 1990 to over 20 % below the average of these other countries in recent years, or almost a 30% greater fall in US rates. American death rates from breast cancer declined from about 10% above the average of these other countries in 1990 to slightly lower.
These results suggest that the US health care system does deliver better control over serious diseases than systems in other advanced countries. Of course, American health care delivery is much more expensive, so a natural question would be whether the greater apparent benefits are sufficient to justify the greater cost?
To get a very rough answer to this question, suppose generously that the American health care system adds 1 life year on average to persons above age 50 compared to what they would have with the average health care system in the 15 comparison countries used by Preston and Hu. Suppose also that people over age 50 value each additional life year by $120,00- since this is a ballpark figure often used for the average American, the dollar value may be lower (or higher!) for older persons. Given that about 4 million Americans reach age 50 each year, the aggregate value placed on these additional life years with these assumptions would be close to $500 billion. This is a little over 4% of American GDP, so this assumed improvement in mortality rates, even aside from improvements in the quality of life, could justify much of the additional spending by the US on health care compared to other wealthy countries.
Of course, the assumption that the American health system produces one additional life year for each person over age 50 may be much too generous, and perhaps older people place a much smaller value on an additional year than $120,000. Still, these calculations suggest that America should hesitate without additional evidence of the type I have used before jumping on the European bandwagon, and conducting radical surgery on the American health care delivery system.

Posted by Gary Becker at 9:13 PM | Comments (32) | TrackBack (0)

American Health Care--Posner's Comment

Last December, the McKinsey consulting firm published a report which states that despite the much higher per capita spending on health care in the U.S. compared with peer countries, the longevity of Americans (even if only that of white Americans is considered) is lower than the average of the comparison countries. This is true, according to McKinsey, even though the prevalence of disease is less in the United States than in those countries (with the principal exception of diabetes, a consequence of Americans' obesity). Because Americans smoke less than the people in those countries, smoking-related diseases are actually lower in the United States.

The report attributes to the higher cost of health care in the United States to higher physician incomes, physicans' control over the number of medical procedures and their ownership of testing and other facilities, which drives up utilization, much higher prices for procedures, higher drug prices, and other factors. To which should be added the exemption of employer-provided health benefits from employees' income tax and the very high overhead costs of health insurers.

Against this, the Preston-Ho article that Becker summarizes points out that the more extensive screening and aggressive treatment of selected cancers, notably breast cancer and prostate cancer, in the United States result in lower mortality from those cancers than in the peer countries.

That is an important point, but it does not establish the superiority of our health-care system. To establish (or refute) that superiority would require conducting a cost-benefit analysis. I have my doubts that such an analysis would vindicate the U.S. system. We spend some $2.5 trillion a year on health care. Our peer countries spend about 60 percent as much per capita on health care and this implies that if we spent at the same level as they, our annual health-care expenditures would be $1.5 trillion. The question, therefore, is what benefits are we obtaining for the additional $1 trillion that we are spending? Suppose the additional screening for and treatment of cancer that we do compared to what the peer countries do is $100 million a year (I have not been able to find an estimate of that cost); that would leave $900 million in "excess" health-care expenditures to explain.

A related point is that the causes of the lesser emphasis in the peer countries on cancer screening and treatment have not been explained. Is it simply a lack of money? Or is it a medical judgment? There is some skepticism in medical circles concerning the overall efficacy both of mammography and of screening for and treatments of prostate cancer. Treatments for prostate cancer are expensive in dollar terms but more so in side effects, which often are permanent. Different people, and perhaps different populations, make different tradeoffs among the various factors that affect a decision on screening and treatment.

I also question the Preston-Ho suggestion that the shorter average life span in the United States compared to that in the peer countries should be treated as a completely exogenous factor. Treating it as such results from an artificial distinction between medical care and public health. Obesity is not a disease, but it is a serious public health problem. A rational allocation of health-care resources might require a shift in resources from end-of-life medical treatments to preventing obesity. Such a shift might increase longevity much more cheaply and effectively than more screening for cancer. So might greater efforts to reduce the murder rate, improve prenatal and infant medical care, reduce speed limits, reduce unsafe sex, increase liquor and cigarette taxes, improve education, reduce poverty, and prohibit motorcycles.

It might be argued that the additional costs of health care that are created by obesity have an offsetting benefit: they reduce the cost of being obese and so increase the net benefits of heavy eating. But the higher health costs of the obese are externalized, in part anyway, to the taxpayer (also to the other members of their insurancce risk pool, if health insurance companies aren't allowed to discriminate). I doubt, moreover, that the obese gain more in enjoyment of food than they lose in the health and other costs of being obese. Much obesity is a result of ignorance (both of calories and of the health effects of obesity), bad habits picked up from parents and peers, negligent parenting, and poor impulse control (i.e., very high discount rates).

And speaking of obesity, its prevalence in the United States undermines studies that find that people attach great value to small improvements in quality and quantity of life. The fact that so many Americans eat badly, don't exercise, drink (or "text") when they drive, and otherwise endanger their life and health, implies, since one can eat well, drive sober, and exercise, etc., at relatively low cost, that people don't value small improvements in quality and quantity of life very much--unless the improvements are paid for by someone else!

Even if we are receiving $1 trillion in benefits from the "extra" $1 trillion that we paying for medical care, it doesn't follow that the $1 trillion in extra costs isn't too much. The reason is that we face, in my opinion, a fiscal crisis; something will have to give and maybe it should be some medical care. The national debt this year will almost equal the Gross Domestic Product (true, the "public" debt--debt owed to entities outside the federal government--is lower than the overall national debt, but the debt owed the social security trust fund, for example, is a real measure of likely future fiscal obligations), and it will continue to soar at least until the economy, and with it federal tax revenues, recover. But it probably it will soar beyond that because the Bush Administration established a precedent of $500 billion annual federal budget deficits that the Obama Administration will follow and probably raise. The health-care reform wending its way through Congress will expand benefits without, it now appears, controlling costs. It is a misfortune that Congress didn't begin with trying to control costs, and then consider whether the nation can afford to expand benefits.

Posted by Richard Posner at 8:15 PM | Comments (44) | TrackBack (0)

July 19, 2009

The Drop in University Endowments and What to Do about It--Posner

College and university endowments have taken a big hit from the drop in the stock market and other asset markets. A drop of 20 to 30 percent is common, and there is suspicion that endowments that contain a significant proportion of assets that are not traded on organized markets, such as real estate, have dropped even more, but without "marking to market" their nonfinancial assets.

The effect of a drop in the market value of endowment on a college's or university's finances depends on a variety of factors. I will give a hypothetical example that may help in understanding the issue. Suppose X College has an endowment that before the crash had a market value of $1 billion. In normal times X cashes 5 percent of the endowment each year to contribute to X's budget, 5 percent being a widely used estimate of the average real return of a typical university investment portfolio. Suppose X's total annual budget is $200 million, with a quarter contributed by the income on the endowment (5 percent of $1 billion = $50 million), a quarter by alumni gifts (apart from gifts intended to become part of the endowment), and a half by tuition. In the economic downturn, I'm assuming, the endowment has fallen by 30 percent, to $700 million; tuition net of financial aid has dropped by 5 percent (because of inability of parents to pay tuition, as a result of declines in their income and wealth); and alumni gifts have (for the same reason) fallen by 10 percent. Then College X's income will have declined by 12.5 percent, from $200 million to $175 million, assuming the college continues to treat 5 percent of the (shrunken) endowment as income.

What should X do (other than expelling its suddenly impecunious students and replacing them with affluent ones of less academic promise)? The typical response has been to cut spending by the amount of the drop in income: in my example, that would require X to reduce its spending by 12.5 percent, which it can do in various ways, the usual ones being laying off staff, freezing hiring of faculty and staff, delaying construction, deferring maintenance, reducing staff salaries, and curtailing extracurricular activities.

At first glance, this seems a puzzling response. Why all this dislocation, instead of either spending capital (that is, taking more than 5 percent out of the endowment) or borrowing? Take borrowing first. Unless a dollar is worth less to College X this year than it will be next year (or whenever its income returns to its normal level), why should it spend less when it can spend the same, with modest effect on future spending, by borrowing (in my example, $25 million)? Lending and borrowing are methods by which the marginal utility of income can be equalized across time when total income varies from year to year. Harvard is borrowing more than $2.5 billion, but it is still cutting its budget by 10 percent.

Similarly, although "spending capital" is a familiar example of improvidence, it can make perfectly good sense. In my example, College X is short only $25 million, and if it takes that out of the endowment, the endowment will fall only from $700 million to $675 million. A problem may be that donors of endowment money may have placed limitations on spending, fearing that a gift that they intended to be perpetual would be eaten up if X dipped into the principal of the gift. But not all endowment money is restricted in this way and trustees of trust funds usually have authority to dip into principal in the event of an emergency. However, X may fear that if it does that it will discourage future contributions to the endowment.

Given that last concern, borrowing seems the superior alternative to spending capital, and yet most colleges and universities seem reluctant to use borrowing to fill the entire gap between income and expenditure; otherwise they wouldn't be cutting spending. Granted, credit remains tight, but the elite universities, at least, are fiscally sound and have alumni in positions of influence in the credit industry.

Maybe the reason the colleges and universities are not borrowing more is that they do not expect their income to recover in the foreseeable future. Nonelite colleges depend heavily on state aid, which is likely to be meager for a number of years--state budgets are in terrible shape. And they draw students from the segment of the population that has been hardest hit by the economic downturn. Elite colleges and universities depend heavily on federal research grants, which may diminish as the government tries desperately to control the rapidly mounting national debt, and on donations by wealthy alumni. Many of those alumni have suffered a permanent reduction in their wealth, and many others are facing the increasingly likely prospect of having to pay higher income taxes, which will make it more difficult for them to pay their kids' tuition. Elite universities may have to limit tuition if they want to attract the best students.

A protracted economic crisis has different effects on different industries, because different industries have different vulnerabilities. The current crisis is speeding newspapers, and print media more broadly, to an early grave, and may yet destroy all three Detroit automobile manufacturers. It will not do in the colleges and universities, but it may have a lingering adverse effect on them that may explain and justify immediate measures to reduce expenditures.

Posted by Richard Posner at 3:04 PM | Comments (19) | TrackBack (0)

How Should Universities React to the Decline in Their Endowments? Becker

The values of most university endowments have taken large hits during this financial crisis. The average decline since the real estate and stock market crashes began is probably over 20%, while the value of some endowments dropped by much more than that. Universities reacted to this severe shock with panic, and they often reduced their spending by too much.
I say "by too much" not because of any confidence that stock markets and real estate will return any time soon to their peak values. This is highly unlikely for real estate, and dubious for stock markets. My reason for objecting to large cuts in university spending is that they have usually under spent relative to their endowments. The philosophy behind these spending rates is that universities should not live off of capital, so that they can pass their capital intact on to future students and faculty. In order to preserve endowment values for the future, universities have tried to spend from their endowments only the income yielded, including capital gains, adjusted for year-to-year fluctuations in returns, and for other risks. On average, they have spent between 4% and 6% of their endowments.
This philosophy has not been implemented, if by not living off capital is meant that endowment values would be held rather constant in the long run. For endowments at all the major universities, and many other schools as well, have grown at very good rates during the past 40 years rather than being maintained intact. For example, Harvard's endowment increased many fold from 1960 to its peak at well over $30 billion in the fall of 2008. The University of Chicago was much less successful in its investment and gift-raising strategies than Harvard, but even the value of Chicago's endowment more than trebled from 1960 to its peak in 2008.
University spending has been too little to maintain endowment values constant partly because the annual return on their endowments has been underestimated, and partly because of large annual gifts from alumni and foundations that raised endowments. These gifts were especially big during the past decade as stocks soared, and as financial executives and others received generous bonuses and stock options, but they have not been negligible even during more normal times.
If we conservatively assume that gifts double endowments over 40 years, than in order to account for gifts and maintain endowment values, schools should spend 1 ¾ percent of their endowments in addition to the total incomes yielded by the endowment. Since incomes have averaged over long time periods at about 5-6% of endowments, the additional spending that would tend to keep endowments constant would raise endowment spending by more than one third. This means a very a large increase in total spending for schools like Harvard that get a big fraction of their annual revenue from endowment income. The percentage increase in spending would be significant but smaller for schools like Chicago that are less well endowed, and get a larger fraction of their revenue from tuition and grants.
More fundamentally, the argument that universities should try to maintain a constant, or "sustainable", endowment value is flawed and not compelling for reasons similar to the criticisms of proposals to maintain an economy's capital over time to achieve "sustainable" economic development. Why should schools aim to maintain endowments constant when even aside from private gifts, endowment income provides only a fraction of their annual revenue? Moreover, technological improvements in the efficiency of spending, by schools, such as more effective use of internet learning, may allow smaller endowments in the future to achieve as much in educating students and conducting research as larger endowments do now.
Another reason for spending more out of endowments than the income yielded is that effective spending on training of students and research, and often also spending on sports, are frequently productive in stimulating greater gifts and governmental grants. For example, schools that produce pioneering research, or that attract able and ambitious students who go on to achieve great success, tend to get larger amounts of foundation and other gifts through the favorable publicity they receive. In effect, spending is productive not only in raising student and faculty achievements, but also indirectly in inducing greater gifts that can lead to even greater accomplishments in the future.
In recent years members of Congress have proposed forcing universities to spend a larger fraction of their endowments, so that endowments do not increase as rapidly as in the past. It would not be wise for Congress to get involved in university spending rates, but for the reasons I have given, it is in the self-interest of well-run universities to spend at much higher levels than they have been doing during the past several decades.

Posted by Gary Becker at 2:53 PM | Comments (17) | TrackBack (0)

July 12, 2009

Legislation on Clean Energy-Becker

In late June the House of Representatives approved The American Clean Energy and Security Act. If the Senate approves a similar version, this would constitute the most important American legislation on overall control of carbon-emitting gases. The main provision of the bill is a cap and trade system, to begin in 2012, which would provide allowances for emissions of carbon dioxide and other greenhouse gases. The goal of the bill is to reduce the carbon emitted by American industries to 17 % below 2005 levels by the year 2020, and to reach more than 83% below 2005 levels by 2050.

Some environmentalists have criticized the 2020 energy-reduction goal as too little and too late. However, I believe that the optimal greenhouse gas policy is to go slow initially until greater evidence on the severity of global warming becomes more apparent. The main threat to the world from global warming is an as yet unknown probability of quite severe warming that would cause considerable harm-the world could adjust at relatively little cost to a moderate degree of global warming. The additional evidence accrued during the next decade will provide more information about the likelihood of the severe warming that would merit more drastic steps. If such steps become warranted, then the rate of carbon reduction and carbon storage should be speeded up beyond that envisioned in the House bill. On the other hand, if milder versions look likely, the 2050 goal of a more than 80% reduction in carbon emissions could be relaxed.

Another reason for going slowly at first is to determine how much will be done on global warming not by EuropThese and other developing countries, along with the US, will be the major contributors to greenhouse gas emissions during the next decade. If the BRICs cannot be bribed or threatened into taking steps to reduce their carbon emissions, the US might rethink how much it wants to do. Rethinking American policy would be especially urgent if the more the US did, the greater the migration of industries from America to developing countries.

While accumulating information during the coming decade on the severity of the global warming problem, the US should greatly invest in trying to achieve breakthrough technologies in advanced carbon control and storage. The aim would be to acquire technological knowledge that could be quickly implemented without enormous cost if the evidence warranted imposing major carbon controls and storage in a short period of time. The House bill allocates about $1 billion annually to the Carbon Storage Research Corp for further research. This is probably not enough, given the possible need to act quickly and decisively to combat global warming.

Under the House bill during the first decade or so, almost all carbon allowances will be given away, mainly to companies, rather than sold to the highest bidders. Over time the fraction sold would continue to increase until the vast majority of allowances would be sold. Many economists have criticized this giving away of allowances during the next couple of decades as a missed opportunity to raise revenue for the federal government through the sale of allowances. In light of the pending massive federal deficits during the next several years, auctions might seem the best approach.

However, the political reality is that significant cap and trade legislation might not gain enough political support if the government sold energy emission allowances rather than giving the majority to the industries most affected. For energy-intensive industries are well organized politically, and they would strongly oppose a carbon tax-which is what an auction of emission allowances amounts to- since such a tax would reduce profits in these industries. On the other hand, energy-intensive industries might support, or only weakly oppose, a cap and trade system where most allowances were given to them since that system could increase their profits, or only reduce them by a little.

Economists typically assume that when a new tax, like a carbon tax, is introduced, government spending is held fixed, so that other taxes can be reduced. That assumption is often useful for analysis, but may not be realistic politically. The revenue from a new tax may be mainly used to increase government spending rather than to reduce other taxes. The case for selling emissions through auctions rather than giving them away is a lot weaker if the government wasted much of the additional revenue, or if the additional spending itself distorts behavior by households and firms. Empirically, the most common response to "new" tax sources, like a carbon tax, is a combination of reduced other taxes and greater government spending (see Becker, Gary S., and Casey B. Mulligan, "Deadweight Costs and the Size of Government." Journal of Law and Economics, October 2003). This typical response makes the case for selling cap and trade allowances considerably weaker than if government spending were held fixed after new tax revenues were collected.

Posted by Gary Becker at 8:07 PM | Comments (71) | TrackBack (0)

The Cap and Trade Carbon-Emissions Bill--Posner

The energy bill just passed by the House of Representatives and awaiting action in the Senate is extremely long and complex, and cap and trade is only one part of it; but like Becker I will confine my remarks to that part, and thus treat the cap and trade component as if it were a separate bill. I will also assume that it will be enacted, and in much its present form.

Becker's analysis of the political realities surrounding the bill are persuasive; and he probably is also right that revenues generated by a tax (in lieu of a quota) approach to carbon emissions would be dissipated on other government programs, such as health-care reform, rather than used to pay off some of the nation's mounting public debt. Moreover, imposing heavy new taxes in the midst of a depression would retard economic recovery.

In principle, a stiff tax on carbon emissions (and other greenhouse gases, but for the moment I'll confine myself to carbon) is, it seems to me, superior to the quota (cap and trade) approach. (I develop this argument in my book Catastrophe: Risk and Response [2004].) Not because it would necessarily reduce carbon emissions more than a quota approach would do, but because it would stimulate research into ways of solving the global-warming problem technologically. The higher costs of energy to energy producers would create strong incentives to develop technologies that would solve the problem, including technologies for removing carbon dioxide and other greenhouse gases from the atmosphere, which may well be a more promising approach than trying to induce the substitution of "clean" energy sources for fossil fuels. (To create incentives for developing technologies for taking carbon dioxide out of the atmosphere, a carbon-emissions tax would have to be complemented by a negative tax--a bounty--for carbon dioxide removed from the atmosphere.)

The incentive effect of the cap and trade bill is weaker. Each energy producer will receive a quota, and many of the producers will be within their quota or able to meet it at low cost. Those producers that cannot comply with their quota may be able to purchase the rights of other producers (that is the "trade" component of cap and trade) at modest cost; for the aggregate reduction in carbon emissions required by the bill--a 17 percent reduction over the 2005 level, phased in between 2012 and 2020--is modest. Moreover, they may be able to buy additional allowances from the federal government (which is holding some allowances in inventory, as it were) at a modest price (depending on what the government decides to charge), or--what looks like a potentially huge loophole--"offsets" to the emissions that they cause. This means undertaking or financing projects to reduce emissions from other sources, such as afforestation projects designed to increase the absorption of carbon dioxide (trees consume more carbon dioxide than they produce). The evaluation of such projects will be extremely difficult.

Assuming the "cap" component of the bill will reduce the output of energy generated by burning fossil fuels, energy prices will rise and consumption therefore fall. The reduction in output may increase the profits of the energy companies, just as when competitors form a cartel to increase prices and profits by reducing output.

True, a quota that reduced the output of an energy producer by the same amount as a tax would create an opportunity cost equal to the tax: that is, the same innovation that would reduce the tax to zero by eliminating the producer's carbon generation would increase the producer's output to its former level. But is a difference in the likely efficacy of the two methods, quite apart from the fact that the tax (to the extent not reduced to zero as a result of an innovation) but not the quota would generate government revenues. As a matter of practice though not of theory, firms often do not react to opportunity cost as quickly as they do to an out-of-pocket cost. The out-of-pocket cost shows up right away on the balance sheet (some of it at any rate--some part will be passed forward in the form of a higher price) and is likely to affect the price of the corporation's stock more quickly than a failure to take advantage of an opportunity to eliminate the cost by innovating.

The effect of the cap and trade bill on the amount of carbon dioxide and other greenhouse gases in the atmosphere is thus likely to be slight, and the administrative costs of the program will be great. Emissions will continue to increase, probably at no lower rate than at present; for the modest effect of the bill will be offset by the growing emissions by China, India, and other rapidly developing countries. Conceivably the bill will provide some impetus to effective international cooperation to limit global warning, but one cannot be very optimistic. Hence the importance of a technological fix, which does not require international cooperation to be effective. If technology were developed for removing carbon dioxide from the atmosphere, Third World countries could emit carbon dioxide to their hearts' content.

We may indeed already have the technological fix, though mysteriously it receives little attention. Sulphur dioxide, the cause of acid rain and the poster child for cap and trade--because the cap and trade program for sulphur dioxide has been a big success--is the opposite of a greenhouse gas: it cools the atmosphere by reducing the amount of sunlight that reaches the earth's surface. Injecting relatively small quantities of sulphur dioxide into the atmosphere would offset the effect of atmospheric carbon dioxide in heating the earth's surface. The opposition of environmentalists to using a pollutant to combat global warming and therefore seeming to approve of pollution, and concern with the bad effects of increasing the amount of sulphur dioxide in the atmosphere (effects that might not be limited to a modest increase in the amount of acid rain), have thus far kept this option from serious consideration in political circles.

The possibility of serious adverse feedbacks makes both economic depressions and climate change extremely dangerous events, warranting emphasis on preventive measures taken well in advance. That is an argument for more aggressive measures than contemplated by the cap and trade bill. But the power of special interests and our soaring national debt make the argument academic.

But may we at least have a decade before the danger of acute global warming becomes acute? Probably, though no one can say for sure. Still, a wait and see approach for a decade is certainly a defensible option.

Posted by Richard Posner at 7:27 PM | Comments (38) | TrackBack (0)

July 5, 2009

The Senate and the Filibuster--Posner

The U.S. Senate is a very peculiar institution. It was not when it started. It was created to be a check on the popularly elected House of Representatives, but also on the President, through its "advise and consent" power--the President's nominees for officials had to be confirmed by the Senate. Senators were not to be elected, but to be appointed by the state legislatures. The assumption was that the legislatures would want to appoint a person of distinction to represent the state. There were only 13 states (though more were envisioned), so there would be only (at the outset) 26 Senators. Their long terms (six years) would encourage expertise and a greater independence from popular passions than the members of the House of Representatives, elected for only two-year terms, could be expected to have. The Senate would, in short, be an elite, deliberative, and only indirectly democratic body.

The change in the character of the Senate since the Constitution of 1787 has been profound. Senators, as a result of a constitutional amendment, are now directly elected, and there are 100 of them. The combination of the amount of time that they have to spend raising money and tending to constituents, and the immensely greater populations of most states now compared to the eighteenth century, and the enormously greater size and complexity of the federal government, have resulted in Senators' being underspecialized, despite having large staffs. The filibuster (a creature of Senate rule rather than of the Constitution or a statute) creates a requirement of a supermajority to pass legislation to which there is substantial senatorial opposition, and rules or customs of senatorial "courtesy" give individual Senators considerable blocking power, for example power to delay confirmation hearings.

The result is that the Senate is an extremely inefficient institution compared to the House of Representatives, in which the majority is in firm command. And because there are so many more House members (435), they have fewer committee assignments and thus can develop greater expertise than Senators; in addition, although they run for office three times as often, they run in much smaller districts and often with little competition and on both accounts don't have to raise as much money in campaign donations as Senators do.

Since the Senate is very large and Senators are directly elected, it is unclear why there is a Senate--that is, why the federal legislature is bicameral. Bicameralism increases the transaction costs of enacting legislation, which can be good or bad (it is bad in national emergencies, as in the financial crisis of last September), and it also increases the cost of repeal, which on balance probably is bad, arbitrarily enhances the political power of sparsely populated states, results in many unprincipled and confusing legislative compromises, and diffuses responsibility for legislation. It is not clear that on balance we are better off with the bicameral system.

The filibuster is an incomprehensible device of government. A supermajority rule, whether it is the rule of unanimity in criminal jury trials or the supermajority rules for amending the Constitution, makes sense when the cost of a false positive (convicting an innocent person, or making an unsound amendment to the Constitution) substantially exceeds the cost of a false negative. But it is hard to see the applicability of that principle to Senate voting, given the other barriers to enacting legislation.

These reflections on the filibuster are prompted by the Democratic Party's recent achievement of a filibuster-proof (60-40) majority in the Senate. It might seem that since the President Obama and Vice-President Biden made such a strenuous effort to convert Senator Arlen Spector from the Republican to the Democratic Party, which put the latter within one vote of a filibuster-proof majority, they must think that having such a majority is a political asset. I am not so sure they do think that. It is easy to see why the conversion was in the Democratic Party's interest regardless of its potential effect on the filibuster; by eliminating one of the Republican Party's most prominent moderates, it contributed to the growing marginality of that party, as it becomes increasingly identified with a rather shrunken right wing.

I am not sure the filibuster-proof majority is a boon to the Democratic Party and program. Because if now the Administration's legislative program fails of passage or is mutilated in the course of passage, it will not be possible to blame an obstructive minority consiting of filibustering Republican windbags. Furthermore, the new voting alignment increases the power of every Democratic Senator, by threatening to align himself with the 40 Republicans in a crucial legislative showdown, to thwart the Administration's program or, more realistically, to insist on what may be costly compensation in the form of an amendment favoring an interest group that is important to his electoral prospects. Indeed, each of the Democratic Senators now has an incentive to play the hold-out in order to extract concessions, in any situation in which Republicans need only one or a very few Democratic defectors in order to defeat an Administration bill.

Posted by Richard Posner at 12:00 PM | Comments (81) | TrackBack (1)

The Senate Filibuster -Becker

The United States, as the name indicates, was formed as a confederation of independent states-initially only 13, but since expanded to 50. The first Federal government was based on The Articles of Confederation, but that was abandoned because it gave too little power to the Federal government. It was replaced in 1787 by the Constitution. Many governance rules have been a compromise between the growing power of the federal government, and the original vision of the country as a confederation of states that retain much of the legislative powers.

Posner describes the evolution of the Senate from a few members appointed by state legislatures to a much larger body elected directly by the voters of each state, where each member serves a 6-year term that is renewable indefinitely. Small states, like Rhode Island, have the same number of senators as the most populous states of California and New York. The Senate is an anachronistic byproduct of the original concept of a federation of states since in the modern world the federal government dominates state governments. The potential influence of a relatively small number of states is increased by the supermajority rule that is embodied in the filibuster.

The Senate filibuster is not a part of the Constitution, but is a rule the Senate established. However, the protection the filibuster gives to minorities does fit in with the belief of the founders of the nation that the power of majorities needs to be constrained in order to protect the interests of minorities. The Constitution, and the checks and balances among the executive, legislative, and judicial branches of government, are important ways they devised to rein in the power of majorities.

Whether someone likes or dislikes the use of the filibuster on particular issues usually comes down to whether they like the legislation that the majority is trying to pass. However, the evaluation of the Senate filibuster as a legislative rule should depend on whether filibusters have blocked desirable legislation more than they prevented the abuse of power by a majority, especially power by a temporary majority. Its main use during the past 70 years is not reassuring since Southern senators filibustered on several occasions against civil rights legislation in order to protect discrimination by Southern states against blacks. However, their filibusters did no more than delay the passage of this legislation, and the delay gave time for the South to become reconciled to the necessity of giving Southern blacks more equal rights.

The mere threat of a filibuster has sometimes discouraged the bringing up of bills in Congress, although these threats also probably only temporarily delayed passage of legislation with a strong and sustained majority interest. Even senators that oppose particular legislation hesitate to filibuster legislation that has a strong mandate since they risk becoming unpopular. Yet they may use the threat of a filibuster to extract a compromise more favorable to their position.

The use of the filibuster has been rising over time: the 110th Congress had over 100 closure votes to try to cut off further debate. The Democratic majority in the present Congress seemed intent on achieving a "filibuster proof" majority of 60 members. Yet, as Posner discusses, having just the minimum number of senators that can enforce the closure rule gives considerable bargaining power to Democratic senators who do not strongly endorse the Democratic position on particular legislation.

I generally support requiring super majorities in the legislature on a selected number of major issues because I believe the abuse of majority power is a greater danger in the world of big government than is the blocking of the majority's will. To be sure, the Senate filibuster is not the ideal way to do this, partly because it is not confined to major issues, and partly because the Senate is based on representation by states rather than by population. Nevertheless, the filibuster may well be better than the alternative of having just straight majority voting in both houses.

Posted by Gary Becker at 11:54 AM | Comments (74) | TrackBack (0)

Sun	Mon	Tue	Wed	Thu	Fri	Sat
�	�	�	�	�	�	1
2	3	4	5	6	7	8
9	10	11	12	13	14	15
16	17	18	19	20	21	22
23	24	25	26	27	28	29
30	31	�	�	�	�	�

Legal Blawgs Web Archive Collection