Becker is right to point out the difference in supply conditions between oil (and other minerals, but I will limit my discussion to oil) and agricultural products: it is cheaper to expand output of the latter than of the former. Hence as demand for oil and for food rise as a function of population growth (an important qualification, as I'll explain--population growth is not the only driver of increased demand for food), oil prices will rise faster than food prices. This is fortunate because while there are substitutes for oil, there are no substitutes for food.

A continued increase in world population will increase the demand for both oil and food, and historical experience suggests, as Becker explains, that the increased demand for food can be met at only modestly increased cost even if the world's population expands greatly, though this depends in part on how rapid the expansion is--the more rapid it is and hence the steeper the increase in the demand for food, the higher the cost of meeting that demand will be, as it is easier to increase production in the long run than in the short run. Moreover, a sizable expansion in population would raise the price of farmland by increasing its opportunity cost.

As the world grows wealthier, the rate of expansion of population should, if historical experience is a guide, slow. But even if population stopped growing altogether, the demand for food would continue to rise because more people (perhaps billions more) would be able to afford the rich diet that people in wealthy countries consume. Supplying that rich diet is very costly in agricultural resources, for one of the major components of the diet is meat and the production of meat requires more agricultural output than the production of cereals and vegetables, since the animals that people eat are big consumers of food.

Technological innovations may hold down increases in the price of food that are due to the increased demand for a rich diet as multiplied by increase in population. But those innovations may create substantial externalities even if they do not push up prices (indeed, the less the increase in prices, the greater the output of agricultural commodities and hence the greater the externalities). As more and more countries adopt the most efficient methods of agricultural production, and thus for example converge on the optimally genetically modified variants of crops, genetic diversity will decline, which will increase the potential damage from blights. (It is not only stock portfolios that benefit from diversification.) Agriculture is a heavy user of water, moreover, and global warming appears to be reducing the supply of water usable for irrigation by reducing the size of glaciers. The run off from the seasonal melting of glaciers provides a more usable supply of water than rainfall, because the water from a melting glacier is channeled, while rain that falls outside a river or other body of water is difficult to store for use in irrigation.

I am one of those timid souls who worry about the downside of technological advance and economic growth. I find the prospect of continued increases in population and income, and of the technological innovations necessary to cope with those trends, unsettling.

The New York Times published an article last Thursday on the Swiss health care system, which can be viewed here: www.nytimes.com/2009/10/01/health/policy/01swiss.html?_r=1&em. The system is simple. There is no "public option," that is, there is no government health insurance program, such as Medicare or Medicaid. There is very little employer-provided health insurance, presumably because employee health benefits are not tax exempt; almost all health insurance is therefore bought by the insured. Everyone is required to buy a health insurance policy that provides a specified minimum of benefits (they can buy more expensive policies if they want), but there are subsidies for people for whom the expense would be a hardship; about 30 percent of the population receives a subsidy. Because of the heavy subsidization, the prices charged by the insurance companies are limited by government, but at a high level. (The limits therefore limit doctors' fees and incomes, and doctors are less well paid in Switzerland, relative to average wages, than in the United States.) There are many insurance companies, and people can switch freely among them. Copayments or deductibles are larger, and as a result the average out-of-pocket cost of health care is higher in Switzerland than the United States--an average of $1,350 per year, versus $890 in the United States. But the aggregate cost of health care is much lower in Switzerland--11 percent of GDP versus our 16 percent--though higher than in any other country besides the United States.

There is, as I said, no special program for the elderly, corresponding to Medicare--which may be why male life expectancy at age 65 is higher in the United States than in Switzerland, although female life expectancy at age 65 is higher in Switzerland and life expectancy at birth is substantially higher in Switzerland, in part because infant mortality is only about half as great as here. The quality of medical care does not appear to be inferior in Switzerland to that in the United States, and there appears to be no problem of queuing, as in Britain and Canada. Indeed the Swiss have significantly more doctors, nurses, and hospital beds per capita than the United States, which suggests that there may be less queuing there than here; and there is general satisfaction among the Swiss with their system, although there is some grumbling over the high cost of medical care.

Of course one must not put too much weight on a single article, but the information in the Times piece appears to be corroborated, at least the statistical data; and some of my description of the Swiss system is drawn from other sources.

If the United States could reduce its medical costs from 16 percent of GDP to 11 percent, the savings would be $700 billion a year; and if the reduction did not reduce the health or longevity of the American population or create queuing costs, there would be no offsetting cost; the $700 billion in savings would be net.

But while the Swiss health-care system may be great for the Swiss, comparing the health-care systems of two countries, even if they are broadly similar (both the United States and Switzerland are wealthy, modern, Western, democratic, capitalist nations), is treacherous, because beneath the broad similarities are potentially important relevant differences. Two of particular importance in the present context are, first, that the Swiss are probably healthier than Americans, on average, apart from any superiority of Swiss health care, and, second, that the Swiss probably have lower expectations of health care than Americans.

The Swiss do not have a large "underclass" (corresponding to the residents of our inner cities) that is poor and has a very high murder rate and high infant mortality and a high incidence of AIDS and other diseases. In addition, the Swiss do not have America's obesity problem, which is a source of abnormally high medical costs because of the treatment costs of diabetes and other diseases to which obese people are disproportionately prone.

And the Swiss people in all likelihood do not expect as much medical intervention as Americans too. Europeans tend to be more fatalistic than Americans. They do not share our preoccupation with extending the longevity of very old people, or our exaggerated faith in medical science that leads some of us to describe the death even of a nonagenerian relative as a "medical failure." Nor do they have as great a propensity as we to insist (after researching a disease on the Internet) on receiving medical care beyond what a doctor's professional judgment thinks warranted.

Our expectations regarding medical treatment are connected to our poor health: Americans want both to indulge in an unhealthy but enjoyable life style and live forever, and they try to square the circle by demanding extravagant (by international standards) health care. (I am exaggerating, of course; some of our poor health is due to ignorance rather than to a deliberate choice to substitute medical treatment for healthful living.)

So we might adopt the Swiss system and discover that our aggregate costs of health care had declined little from their current 16 percent of GDP. Indeed, because of increased coverage, it might increase (see below).

The proper use to be made of the experiences of other nations with health care is not advocacy for our adopting the health-care system of a nation broadly comparable to ours that spends a lower fraction of GDP on health care than we do. It is to note the methods used by foreign countries whose health-care systems are well regarded by the local population and see whether any of them could work well here, bearing in mind the dangers of piecemeal adoption of foreign methods. (An example of those dangers is the adoption by the Detroit auto companies some years ago of the "quality circles" used by Japanese auto companies to increase productivity by encouraging their workers to suggest productivity-enhancing innovations. The quality circles failed in Detroit because the auto companies did not realize that what made the quality circles work in Japan was the practice of lifetime employment; our workers were reluctant to suggest productivity improvements because they knew it might well result in a smaller workforce and therefore in layoffs.)

The features of the Swiss health-care system that seem well adapted to American conditions (though whether their adoption would be politically feasible is a separate question--to which the answer is "no," at least at present) are, first, repealing the tax exemption for employer-furnished health benefits, since the exemption both creates an artificial incentive for employers rather than employees to buy health insurance and disguises the cost of the benefits to the employees (in lower wages); second, making everyone buy health insurance, in order to prevent adverse selection (that is, excess demand by the unhealthy), the problem to which group (normally employer-group) insurance is a second-best solution; third, requiring significant copayments or deductibles so that the marginal cost of health care to the insured is not so low as to induce the overuse of medical resources; and fourth, providing no special program for the elderly, but instead requiring them to buy insurance like everyone else, with the cost subsidized only if they cannot afford the cost of the insurance rather than just because they are old.

Such reforms would probably produce a net savings in aggregate U.S. health-care costs, though this is not certain, because of the subsidies and because any extension of coverage--which would be considerable because everyone would be required to have health insurance and the number of uninsured in the United States exceeds 40 million persons--is likely to increase the demand for health care. The subsidies are transfer payments rather than costs in the sense of consuming real resources, but worrisome nevertheless because of the potential long-term harm to the economy from our soaring public debt. But the aggregate transfers and (real) costs would probably be less under a version of the Swiss approach than under the approach urged by the Administration, which does not have credible cost-saving measures build into it.

The Swiss health care system has several important properties that I (and many others) have been advocating should be incorporated into any reform of the US health care system. One major advantage of the Swiss system is that employer-provided health care does not receive any special tax breaks, whereas the US system is built on these tax breaks. As a result, only a rather a small fraction of Swiss health care is obtained through employment. Mainly, Swiss families buy health care on their own, so that, unlike in the US, their health insurance does not reduce their incentives to change jobs because job changes do not endanger their health coverage. Unfortunately, probably due to union pressure, Congress is not planning to eliminate this tax break for employer-provided health care. Indeed, many Congressmen want to increase the pressure on employers to provide health care to their employees.

The Swiss system includes a mandate that everyone buys a minimum amount of health care coverage. This solves the American problem where over 40 million persons have no health care coverage. If uninsured persons get sick-fortunately this is not frequent since they are mainly young-that raises the cost to everyone else since the uninsured typically seek treatment for any illnesses at hospital emergency rooms. The health care reform bills in Congress do include various coverage mandates, although they are not as straightforward or as desirable as the Swiss mandates.

The Swiss system typically has much larger co-payment rates than the US does for anyone seeking medical care or buying drugs. By shifting more of the cost to individuals and away from insurance companies, the Swiss give individuals greater incentives to economize on their health care spending since health care is more expensive to them. On the other hand, the Swiss system does not seem to have the equivalent of health savings accounts (HSAs) that allow consumers to carry over from one year to the next any balances in their health accounts that are not spent. These HSA accounts should become a more important part of the American system.

The Swiss do not give any special medical advantage to older persons, for they have access to the same health subsidies and same private health insurance system, as does everyone else. The US could approach the Swiss way by making Medicare much more means tested, so that higher income older persons would pay a much larger share of the costs of their medical care than they do now. Unfortunately, neither President Obama nor either political party is willing to tamper much with the Medicare system as presently constituted.

The Swiss system has no public insurance option, and relies on competition among private health insurance companies. I have argued strongly against a public option (see my post on August 17th of this year), and while it appears that this option is being dropped from most Congressional bills, liberal Democrats are still lobbying to have such an option included in any reform package.

Although I do not know the details of the Swiss system, it appears to provide good health care while spending only about 11% of its GDP on health care compared to the US' 16%. I say " appears" because previously the British and then the Canadian heath care systems were held up as models for the US to emulate until further evidence revealed that these systems had serious flaws, such as long queues for many types of treatments. In fact, the Swiss system does have some unattractive features that should not be emulated when reforming the American system.

For one thing, the Swiss impose sharp price controls on drugs, lab tests, and other medical procedures. To take drugs as one important example, Swiss price controls reduce prices of top selling US patented prescription drugs to about 40-50% below their American prices. In particular, the cost of lipitor in Switzerland is about 1/3 of its American price. In reality, what the Swiss (and other countries) do is free ride off of the incentive provided by American drug prices for pharmaceutical companies to invest the huge amounts of resources required to produce blockbuster drugs like lipitor.

Very small countries like Switzerland can get away with this free riding since their demand for drugs is so much smaller than that of the US. However, were the US to emulate the Swiss system, and there is a call from some Congressmen for greater control over drug prices, the incentives biotech and pharmaceutical companies have to innovate would be greatly reduced. It is precisely the greater price freedom in the US that induced many drugs companies to relocate their research labs out of Europe and into the United States.

Even though the Swiss spend a much lower fraction of their incomes on health care, their life expectancies at age 50 are about 1.5 years better than those in the US. However, as I argued in earlier posts (see, for example, July 28 of this year), life expectancies depend on many other factors than medical care, and the United States does not look good on most of these factors, such as obesity. More relevant comparisons are access to various tests, such as mammograms and PSA tests, and survival rates from major diseases, such as cancers and cardiovascular disease. The US does much better than other countries, including Switzerland, on both sets of criteria.

So despite the obvious conclusion that various reforms of the American health care delivery system are desirable, Americans are getting some important advantages for their large spending on health care. It is crucial that these advantages not be forgotten when evaluating how much better other countries health delivery systems, including the Swiss system, are than the American system, and in deciding how to improve the American system.

The major American trade unions, including the United Automobile Workers, the United Steelworkers, and the Service Employees International Union, went all out in their support of Barack Obama during the past presidential election. They supplied money-said to exceed $400 million- and hundreds of thousands of volunteers working for Obama. It is believed they were important in his winning industrial states like Michigan, Wisconsin, Ohio, and Pennsylvania.
Naturally, unions expect some payback after the resounding victory of Obama, as all interest groups do when the candidates they support win. Unions are behind the Employee Free Choice Act introduced in Congress early after Obama's election. That Act would make it much easier for unions to be certified to represent the employees of a company. The Act, still bogged down in a divided Congress, would allow for open rather than secret voting on whether a union should represent the employees, and would mandate arbitration over union-management contracts. Because of opposition from some Democrats as well as almost all Congressional Republicans, it is not yet clear how much the legislation that eventually emerges will shift union-management relations in favor of unions.
The bailouts of General Motors and Chrysler have been a second major effort to help unions. In this case, the Obama administration spent tens of billions of dollars of taxpayer revenue to help these companies. I expect total Federal spending on GM and Chrysler will eventually equal or exceed $100 billion. The best alternative to the bailout of these auto companies would have been to allow them to enter bankruptcy proceedings in the fall of 2008 when they were bleeding large losses. After a year or so in bankruptcy they would likely have emerged with considerably lower obligations for health and pension benefits, and reduced hourly earnings of their employees. They might then have competed without additional support against foreign auto companies with plants in the US or abroad.
Instead of allowing such bankruptcy proceedings to occur, and in order to reduce the hit that unionized autoworkers would have to take, GM and Chrysler were bailed out generously, and the federal government in effect became the principal owners of these companies. Although GM and Chrysler were allowed to go into bankruptcy for a short period, in my judgment the main aim of the bailout was to reduce the effect of the financial troubles these companies were having on the earnings and fringe benefits of present and retired autoworkers. Taxpayers paid what the autoworkers should have paid.
Perhaps the most disturbing tilt toward catering to union interests is the very recent 35 percent tariff the US government imposed on imports of Chinese tires without any finding of illegal trade practices by either the Chinese government or Chinese tire manufacturers. The White House under special trade rules can impose punitive measures without finding any "fault". However, this is the first time any president of the United States has used this provision to penalize manufacturers of imported goods or services. It is not only a terrible precedent, but also this may encourage other countries to follow similar procedures, and impose tariffs on US exports that unions and companies in these countries feel are hurting them.
Unions and their allies have succeeded in placing Buy America provisions in the $787 billion stimulus package, against the objections of many foreign countries. Unions are also active in trying to get similar provisions in the cap and trade climate bill that will eventually be passed by Congress and supported by the president. Buy America provisions to stimulate employment of American workers is no different than imposing tariffs to cut imports and increase demand for domestic goods. Both are inconsistent with the goal of free trade, and invite retaliation by other nations.
Politics in democracies allows powerful political interest groups to influence legislation in their favor. In that respect, the steelworkers and other unions are not doing anything differently than, for example, medical insurance companies or auto companies, try to do. But recognizing that this is the way the political process works does not mean that the effects tend to raise the general welfare of the population, as opposed to the welfare of small powerful interest groups.
In this regard, note that non-governmental unions contain only about 8% of the civilian labor force. This means that the benefits they receive from flexing their political muscle under the present White House mainly hurts other workers and consumers. In particular, the tariff on tires will raise the cost of tires to American consumers and make them worse off. Similarly, the bailout of the auto companies will raise taxes and probably also auto prices, and Buy America provisions will make the cost of goods more expensive because they cannot be obtained from cheaper foreign producers. The overall impact of these steps is a less efficient American economy, and substantial harm imposed on American consumers and non-union workers.

Becker is correct that the tariff appears to be a pay back to the unions for their strong support of Obama in the 2008 election. The significance of that support was amplified by a questionable feature of our political system. All but two states award all their electoral votes to the candidate who wins a plurality of the popular vote in the state. This makes winning, however narrowly, the popular vote in states that have a lot of electoral votes disproportionately important to a successful strategy for a presidential candidate, and in turn amplifies the effect of interest groups in those states. States in which unions, despite their modest fraction of the labor force overall, are electorally powerful include major swing states, such as Ohio.

Industry-wide unions are labor cartels, but the aggregate economic effects of unions in the American economy is probably slight. Many unions are unaggressive, being chiefly interested in union dues. Others operate primarily to protect workers against arbitrary supervisors and unsafe working conditions, and these unions may generate net benefits--may even improve labor relations by increasing employee trust. The United Auto Workers is a dying remnant of the dinosaur era of industrial unions; the UAW has done much to bring down the U.S.-owned domestic auto industry but it could not have succeeded had it faced competent management.

Unions reduce management flexibility, and that is particularly harmful in a depression or recession. Union contracts usually require that layoffs be strictly inverse to seniority, so the employer cannot use a depression-generated need to lay off workers to get rid of dead wood; for that reason and because of wage inflexibility created by union contracts, unionized firms cannot cut their costs in response to a fall in demand for their products as rapidly or deeply as nonunionized firms can, and this retards the restoration of economic equilibrium. The effect is particularly pronounced in a deflation, which in fact we are experiencing. In a deflation, constant nominal wages result in increased real wages, thus raising producers' costs and spurring layoffs.

Because unions are quite weak in the nation as a whole, the importance of union support to presidential candidates doesn't necessarily translate into strong support in Congress for pro-union policies. Congress doesn't kow-tow to presidents even when the political party to which the president belongs dominates Congress. The tariff on Chinese tires is a unilateral presidential gift to a union. One hopes it is just a matter of throwing a small bone to the union movement, because to launch a trade war against China would be playing with fire.

There was much more at stake for unions in the auto bankruptcies, but I remain sympathetic to the government's bailouts, especially the initial bailouts of last December. Because of the credit crunch, it is unlikely that bankrupt auto companies could have attracted "DIP" financing--that is, financing the operations of a company that is in bankruptcy but is continuing to operate (such a bankrupt is called a "debtor in possession" (DIP)). Without DIP financing, GM and Chrysler would have had to liquidate (that is, shut down), because they could not operate without credit. Their liquidation would have thrown hundreds of thousands of workers, perhaps more than a million, out of work at a time when the economy was in a steep downward spiral; the effect could have been catastrophic.

By the time the auto companies were allowed to declare bankruptcy, in May, they had undergone a gradual partial liquidation and the panic phase of the economic downturn had passed, so the economy could take the bankruptcy in stride. Still, without government assistance, there was a danger of liquidation and so a case for the further bailouts--though not for the government's becoming the actual owner of an auto company, namely General Motors.

I do not approve of the "Buy America" provisions of the $787 billion stimulus, which appear to be another bone thrown to the unions, because of the danger that they may provoke retaliation. However, they are at least understandable as a stimulus provision. The purpose of a stimulus program is to increase employment, and to the extent that stimulus moneys are used to buy imported goods, the purpose is thwarted--the stimulus then stimulates a foreign economy, not the U.S. economy. Much better than including "Buy America" provisions in the stimulus law, however, because much less likely to provoke retaliation, would be targeting the stimulus expenditures more carefully on the production of goods and services that involve minimum use of imported components, road building being the obvious example though another is military equipment, as emphasized by Martin Feldstein. The stimulus program was in my opinion an essential measure for fighting the looming depression, but it was poorly designed.

We will not be blogging this week. We will next blog next Sunday, September 27.

Vice President Cheney is reported (I don't know whether accurately or not) to have said that "deficits don't matter." Certainly the Bush Administration ran big ones, as a result of which the public debt (which is the national debt less federal liabilities to Americans created by the social security and other entitlement programs) doubled. It has continued mounting as the deficit continues growing, and has now reached $7.5 trillion, which is more than half the (annual) Gross Domestic Product.

It will continue to grow rapidly, because of the fall in federal tax revenues as a result of the economic downturn, because of the aging of the population which along with the continued acquisition of advanced medical technology is causing a continuing rapid increase in Medicare costs, because of the reluctance of Congress to raise taxes or cut any spending programs, and because of the likely cost of the ambitious new programs of the Obama Administration. How much money will actually be appropriated for the programs, such as health-care reform and climate control, is, at this writing, unclear.

The public debt is funded by Treasury borrowing (actually a bit of it is being funded as an emergency measure by the Federal Reserve, but I will ignore that), of which more than 40 percent comes from foreign governments and other foreign entities and the rest from Americans, including banks and other financial institutions. Much of this borrowing is in the form of 10-year Treasury bonds, which are now commanding an interest rate of about three and a third percent.

The government is having no difficulty at present in borrowing at moderate interest rates to fund the public debt, large as it is. The reason is partly that Treasury securities are safe in the sense that there is no risk of default and the current global economic downturn has increased the demand for safe investments, and partly that the world is awash in dollars because of the policy of a number of major nations, such as China (but not only China--others include Germany, Japan, and the oil-exporting small nations of the Middle East), of running very large current-account surpluses (i.e., trade surpluses). These surpluses are largely in dollars because the dollar is the principal international reserve currency, which is to say a currency used in international transactions in preference to using local currencies that fluctuate more than the dollar does. As the world's principal sourcce of international reserve currency, the United States in effect sells dollars to the rest of the world to provide liquidity in international trade, and many of the dollars come back to the United States in the form of investments in Treasury securities, especially from countries that have large dollar reserves because they export much more than they import.

A country that supplies a major international reserve currency must run a current-account deficit because otherwise the rest of the world wouldn't have enough of the currency for their transactions. The fact that foreign countries need large dollar reserves for this purpose means that there are a lot of foreign dollars available for the purchase of U.S. securities, quite apart from current-account surpluses. This makes it easy for the United States to borrow at reasonable interest rates to fund its public debt, even if Americans, unlike Japanese, are not big savers. (The fact that Japanese are big savers enables Japan to fund a public debt that is proportionately much greater than ours, without much difficulty.) Americans are saving much more nowadays than they were a year ago, but this may change as the economy recovers.

As long as Americans are saving a lot, and wanting their savings to be safe, and foreigners as well, and as long as nations like China are running huge current-account surpluses, we can fund our public debt at reasonable interest rates. But that is provided it doesn't grow too fast, and it is growing very fast and there are no signs of its slowing. As the economy recovers, federal tax revenues will rise, but federal expenditures will be rising too, and rising all the faster if a significant part of the Administration's ambitious program is authorized by Congress, because there don't seem to be any serious efforts at either increasing any taxes (even by reducing deductions) or cutting any spending programs. The perfection of interest-group politics seems to have created a situation in which taxes can't be increased, spending programs can't be cut, and new spending is irresistible. Judging by the Bush Administration's profligacy and its impact on the public debt, the situation is bipartisan.

At some point the wheels may start coming off the chassis. Assume that the public debt continues its rapid growth because government spending increases rapidly but Congress refuses to authorize significant increases in taxation. The Treasury will have to borrow more and more, yet at a time when recovering economies need investment capital, forcing interest rates up and hence deepening our deficits. We already pay more than $400 billion a year in interest on the public debt, and that amount will rise rapidly as both the size of the public debt and interest rates rise.

Assume further that political pressures prevent the Federal Reserve from raising interest rates in order to head off inflation caused by the banks' finally deciding to lend (as the economy recovers) the huge excess reserves that they have accumulated as a result of the Fed's open-market operations during the current economic crisis. Fear of inflation will push up long-term interest rates, including rates paid by the Treasury to fund the growing public debt. Fear of inflation will also make foreign countries worry about the value of their dollar reserves, and wonder whether the dollar should continue to be the predominant international reserve currency.

As the dollar falls in value, however, the public debt will become cheaper to repay, the demand for U.S. exports will grow, and our demand for imports will fall. The increase in the ratio of exports to imports will reduce the current-account deficit and thus reduce the rate of increase of the public debt. But increasing exports relative to imports, by tending to reverse the long-term decline in U.S. manufacturing relative to services, may be a painful and protracted one. We have grown accustomed to financing our consumption by borrowing heavily abroad to pay for manufactured imports and for our elaborate systems for distributing goods and providing other services. Our economic productivity has become heavily dependent on the immigration of high-IQ professionals, but one casualty of the current economic crisis has been restrictions on immigration that are designed to protect Americans' jobs.

Moreover, even with a reduced current-account deficit, U.S. public debt will be rising because of increasing unfunded expenditures on medical care and other social programs, and for all one knows on military activity as well since the United States remains the world's policeman. And lenders will charge higher interest rates to continue to fund our public debt if they think the dollar is losing value because of inflation. If inflation persists, then given that there are other international reserve currencies, namely the euro and the yen, and in time the renminbi (the Chinese currency), the dollar will decline as an international reserve currency, and, with the demand for dollars thus reduced, its value will fall further.

As real interest rates rise as a consequence of the growing public debt and decline in demand for the U.S. dollar as an international reserve currency, U.S. savings rates will rise, and by reducing consumption expenditures this will slow economic activity. Economic growth may also fall as more and more resources are poured into keeping elderly people, most of whom are not highly productive members of society from an economic standpoint, alive. The United States may find itself in the kind of downward economic spiral in which "developing" countries often find themselves. As an economic power we may go the way of the British Empire, which occupied approximately the same position in the world economy in the early twentieth century as the United States does today.

Deficits arise when government spending exceeds the revenues raised from various taxes. Deficits add to the stock of government debt. In evaluating the consequences of deficits for an economy, it is first of all crucial to know whether the source of a larger deficit is greater government spending or reduced tax collections, possibly due to reductions in tax rates. The second issue is the burden to the economy of financing the interest payments due on the government debt. I take these issues up in turn.

To the extent that the source of the rise in a deficit is increased government spending, then whether that rise is justified depends on how socially valuable are these government expenditures. By that I mean the social rates of return on these expenditures, such as longer lives for the elderly, relative to the interest cost of raising the required funds, and relative to the returns on other investments in the economy. Much of the increased government spending in different countries during this recession went to help out banks that were in danger of going under. While numerous mistakes were made that will be argued about for a long time, such spending on the whole was necessary in order to limit the financial crisis that had developed.

Other parts of the increase in spending in most countries are far more dubious and may even have harmed their economies. I include in that most of the $800 billion Obama stimulus package, much of which is still not spent even though the brunt of the recession is over This package was promoted as a way to fight the recession, but mainly it is an attempt to reengineer the economy in the directions of larger government favored by many liberal Democrats. I believe much of this reengineering will hurt the functioning of the economy, and of course at the same time will add to the debt burden.
A very small example was the cash for clunkers program in the US that ended a short time ago. The 19th century French essayist Frederic Bastiat discussed facetiously the gain to an economy when a boy breaks the windows of a shopkeeper since that creates work for the glazier to repair them, and the glazier then spends his additional income on food and other consumer goods. The moral of that story is to hire boys to go around breaking windows! The clunkers program was hardly any better than that (see our discussion of the clunkers program on August 24th).

Deficits also arise automatically during recessions because tax revenues fall as the growth in aggregate incomes slows down, and even becomes negative, as it did during this recession. This automatic effect on deficits during recessions from falling tax revenues is supposed to be balanced by automatic surpluses during prosperity times as tax revenues grow because income are expanding faster. Unfortunately, the period prior to the recession also had budget deficits, even though incomes grew quite fast, because Congress and President Bush pushed for greatly expanded spending.

To turn to the second question, will financing the debt become a serious obstacle to rapid US growth as the current and projected sizable US deficits increase the ratio of government debt to American GDP? That depends on four critical variables: the size of the debt/GDP ratio, the level of interest rates, taxz rates, and the rate of productivity growth in the American economy. Suppose the debt held by the public-which excludes government debt held by other government agencies- reaches 100% of present or near term GDP, which is not unlikely.
The burden on the government budget that this imposes depends on the interest rates on the debt. At an average interest rate of 5%, that means 5% of GDP would go to servicing the debt, which is a little less than 20% of total federal government spending. This might be manageable but it is not trivial. On the other hand, if average interest rates were only 3%, servicing costs would be far more tolerable. In fact, the US has been paying about 3% on its debt, so even a considerable increase of the debt to 100% of GDP would still be manageable. But if the Fed starts raising real interest rates to head off the inflation potential in the $800 billion of excess reserves, the debt burden could become a major problem. Another factor is the savings rates coming from the Asian countries, like China. If their savings decline sharply, that too would raise world interest rates and increase the debt burden for all countries.
Rapid productivity growth and an improved tax structure could save the situation because the expansion of GDP caused by such growth and better taxes lower the ratio of the debt to GDP, and makes financing the debt easier. To maintain rapid productivity growth requires that an economy provide powerful incentives to invest in physical and human capital and R&D. It also requires that Congress and other legislatures do not start growing government spending as GDP grows rapidly. But members of Congress and other legislatures are tempted to use much of the growing tax revenue on their pet projects.

One important determinant of the incentives to invest is the tax rates on the rewards of investments in new knowledge and capital. Rich people should pay a larger share of the tax burden, and they do. However, if the emphasis changes from encouraging investments to redistribution, the poor as well as the rich will suffer. Probably the poor will suffer more since the rich can more their capital and themselves to the many low tax jurisdictions in the world.

On October 7, 2008 I wrote an op-ed piece for the Wall Street Journal ("We're Not Headed for a Depression") in which I said there would not be a depression, certainly nothing at all resembling the Great Depression of the 1930s. As the economy continued to decline after that I began to worry that my predictions were going to look foolish, and become famous as one of the many absurdly bad forecasts.
Fortunately for me, and even more so for the world, the forecast turned out to be basically correct. I recently claimed in a post on this blog on August 9th that the current world recession is over, and many economists and official organizations since then have come to the same conclusion. The recession was big and world wide, but it was far from a depression-a rule of thumb is that a contraction is a depression only if the fall in output is at least 10%. The output fall in the US and the world has been less than 5%. Indeed, this recession is hardly more severe in the US- the epicenter of the financial crisis - than a couple of previous recessions, such as the one from 1973-75 brought on by the first oil shock, and the one in 1981-83 resulting from the Fed's successful efforts to squeeze inflation out of the system.
During the Great Depression American unemployment peaked at 25% and was high during the whole of the 1930s, while output declined by more than 20%. During this present (or past) recession, output has fallen by a little over 4%, and unemployment so far has remained under 10%-the latest figure gives an unemployment rate of 9.7%. Since a world of difference exists between the two events, the prevalent fear of a major depression was never realized.
Those who are more pessimistic about this recession point out that unemployment is still rising, and may reach a much higher level than its present rate. They also rightly indicate that total unemployment and underemployment is much higher than 9.7% because some persons have only found part time jobs, while others have been so discouraged by the weak labor market that they quit looking for work, and so are not counted as unemployed.
I will take up both aspects of this pessimism in turn. To understand what has been happening to unemployment, it is crucial to recognize that employment has declined, and unemployment has risen, much more relative to output during this recession than in past recessions because labor productivity-measured by output per worker or per hour of employment- has continued to grow during the recession. Productivity grew by 0.3% during the first quarter of 2009, and by a whopping 1.8% during the second quarter. Typically, measured productivity falls during serious recessions because of excess capacity of capital and the many employed workers who are underutilized. Basic arithmetic indicates that for any given fall in output, the greater the rise in measured labor productivity, the greater the fall in employment, and the greater the increase in unemployment.
Unemployment is typically a lagging indicator in the sense that it usually begins to fall only months after output has started to increase again. Since I expect output to rise only a little in the US during the third quarter that will be over at the end of September, unemployment should continue to rise for a while, almost certainly surpassing 10% at its peak. However, if, as I expect, the growth in productivity will continue into the future at a good pace because of the many innovations and inventions coming on line, that will lead to greater, not a lesser, growth in employment. For at some point, the economics of the positive relation between productivity and employment becomes more powerful than the short-term arithmetic negative relation that occurs during recessions.
In the longer run, advances in productivity are partly produced by investments in R&D and other innovations that generate new products and new processes. Both new products and new production methods typically require investments in both physical and human capital. They also stimulate the use of more workers of various skills that utilize the greater capital stock. This is why over longer time periods, productivity advances and robust labor and capital markets in different economies are strongly positively, not negatively, related. For this reason, the continuing advances in productivity in the US and elsewhere will at first limit and then reverse the falls in employment and rises in unemployment.
It is true that the total underemployment rate during this recession would be well above the official unemployment rate of 9.7%. Some estimates put total underemployment at over 16%, which includes individuals who are reluctantly working only part-time, and also persons who have given up looking for work. However, apples have to be compared with apples, and in judging this recession relative to prior ones, the same calculations have to be made for these past recessions as well. Exactly the same type of growth in underemployment was operating in these prior recessions, and especially for the severe recession of the 1930s. Perhaps the fractions of reluctant part timers and persons who stopped looking for work are greater during the present recession than recessions than say in 1973-75, or 1981-83, but I have not seen any demonstration of this. My guess is that whatever differences exist, they are not enough to reverse the ordering of the severity of different post-war recessions.
My overall conclusion is that productivity advances will lead the world out of the recession, and after a while toward a decent rate of growth in world GDP. These advances will occur even if the financial sector is not fully recovered from its crisis. As productivity advances continue at robust levels, that will stimulate the demand for labor, and begin to reduce unemployment and produce sizable rates of growth in employment.

I am not bold enough to make forecasts about economic recovery, given the unusual economic situation that the country is in. The recovery may be fast or slow, shallow or steep, continuous or interrupted--and if fast and steep may set the stage for inflation and other economic troubles. So I am neither an optimist nor a pessimist.

I am uncomfortable with the way in which modern economists discuss economic downturns. Before the 1930s depression, economic busts were called--"depressions." As far as one can judge from the incomplete nineteenth-century economic statistics, that depression was of unprecedented severity, and hence came to be called the "Great Depression." Which is fine. But thereafter, for reasons I can't fathom, economic busts, instead of being called "depressions" (though of course not "Great Depressions," because they were much less severe), came to be called "recessions." The current downturn, because it is the worst since the Great Depression, is now being called the "Great Recession." I find this lexical nitpicking distracting and unhelpful. Why not just say, we're in a depression, severe by postwar standards but mild compared to the Great Depression?

I also question the convention that says that a depression (or recession, if one insists on retaining that word) ends when GDP growth resumes. Actually, that is not the official (National Bureau of Economic Research) position; its business-cycle committee looks at other factors as well, such as employment. It would be a nonsensical convention applied to the Great Depression; it would imply that the Great Depression ended in March 1933, when output and employment began to rise from their respective one-third and one-quarter decline from 1929.

I would prefer to say that a depression ends either when economic output returns to its pre-depression level or, better, when it returns to the GDP trendline of average annual growth, which is about 3 percent in real terms. So this depression has not ended.

This depression was never likely to be as severe as the Great Depression. One reason is the automatic stabilizers, such as unemployment benefits and other social-welfare programs, and progressive income tax. Another reason is changes in the composition of the workforce. Manufacturing and construction, two of the industries most likely to respond to a fall in demand by laying off workers, account for a much lower percentage of the U.S. workforce today than in the 1930s; and services (which had a low unemployment rate even in the Great Depression) account for a much higher percentage. In addition, there is federal deposit insurance, and a clearer understanding that in a depression the government should try to increase the money supply, and indeed should try to create at least a mild inflation. The Roosevelt Administration did both things as soon as it took office and they probably were responsible for the rapid improvement in the economy that began soon after his inauguration, though it was later interrupted by the economic dive in 1937 and 1938--what has been called the "second depression."

Nevertheless this depression resembles the Great Depression in one respect that makes forecasting particularly chancy--it has been accompanied and made worse by a financial crisis. The normal depression comes about either from something that happens in the nonfinancial economy, such as a big increase in productivity which causes unemployment, or by the action of a nation's central bank in raising interest rates to stop or head off inflation. In both cases, as shown in research by Christina Romer and others, the depression can be effetively treated by the central bank's reducing interest rates, which stimulates economic activity by increasing lending.

But we are in a depression in which interest rates are very low. Indeed, the Federal Reserve is maintaining the federal funds rate at just a shade over zero percent and has been for many months. There are other interest rates, and the effect of the federal funds rates on them is complex, but nevertheless there is nothing further the Fed can do, or at least that it wants to do (because it's beginning to worry about a future inflation), to lower interest rates, though credit remains very tight because the banks remain undercapitalized and demand for loans is weak because people and many businesses are overindebted.

It's because monetary policy, though in combination with bailouts it has saved the banking industry from bankruptcy, cannot do anything to stimulate economic activity that we have the $787 billion stimulus program and other programs, such as the federal subsidies that are keeping GM and Chrysler in business. These programs may be responsible for the recent improvement in the economy, at least in part, though there is no good evidence.

The consumer price index is lower than it was a year ago, which means we're in a deflation. It's a mild deflation, but any deflation increases the burden of debt, which in turn reduces personal consumption expenditures and investment. The unemployment rate is high and rising, and the underemployment rate, 16.8 percent in August, is very high. Housing prices remain very low, which increases indebtedness because a house is the principal asset of most people, and mortgage debt obviously does not fall when the value of the mortgaged property falls. The fall in housing prices, by wiping out the housing equity of milliions of people, exacerbates unemployment by making it more difficult for the unemployed to seek jobs in different parts of the country--they can't afford the down payment on a house if their existing house is worth less than the unpaid balance of their mortgage.

Another factor retarding recovery is the reluctance of older workers to retire, because their retirement savings are impaired. Employers are reluctant to lay them off for fear of being accused of age discrimination, which is illegal. With fewer workers exiting the work force, there is less room for the thousands of people who each day are looking for a job.

There are factors pushing in the opposite direction--toward a rapid recovery. As manufacturers work off their inventories, production restarts; as people's incomes fall, they divert more income to consumption and less to savings; when their incomes fall really far, they start spending their existing savings; and as durables wear out, the demand for durables increases. (It's the fact that the purchase of durable goods is postponable that leads to such drastic falls in manufacturing in a depression, compared to services.) And as economic conditions improve in other countries, U.S. exports will rise, which will stimulate U.S. output.

I don't know how these factors balance out, and I suspect no one knows. After the economists and the businessmen alike were caught by surprise by the housing and credit bubbles and ensuing financial crisis, all macroeconomic forecasts should be treated with a measure of skepticism.