In my post on Sunday I gave an example to illustrate the Treasury's plan to buy the "toxic" assets of banks. Since I left a few quite important implications of the example unclear, this addendum will consider the same example in more detail.
Recall that the Plan would encourage hedge funds and other financial institutions to bid for bank assets. These institutions would only have to put up about 7% of their bid price since the Treasury will supply another 7%, and the FDIC will loan the remaining 86%. If assets appreciate in value over the bid price, the Treasury and funds share the profits equally after the FDIC is repaid. If the asset declines in value, funds are only liable for 7% of the decline, and the FDIC and Treasury absorb the rest.
In my example, there is a 10% chance that an asset will be worth $1000, and a 90% chance that it will be worthless, so that the full expected value of the asset is $100. Assuming competition among funds forces them to bid the expected value of this asset to them, how much will they bid? When the asset pays off $1000, a fund would get half the difference between $1000 and 86% of its bid price, while if the asset becomes worthless they would be compensated for everything but 7% of what they bid. The expected value of these outcomes is approximately $243, and that would be the price that competition forces hedge funds and other funds to pay for the assets. Since the expected value of the asset is only $100, government subsidies would encourage funds to bid about 2 1/2 times the true worth of the asset!
The government would pay the difference between the bid price and the worth of an asset, or $143 in this example. Contrary to many assertions made about the Treasury Plan, this subsidy does not on average go to successful bidders since the expected cost of the asset to them equals the expected value of the asset to them. Of course, the luckier buyers of these assets can make a lot of profits, and they could be subject to Congressional wrath that they profited at the government's expense. The $143 subsidy goes to banks, for they would receive almost 2 1/2 times the worth of their "toxic " assets.
Perhaps good reasons motivate the government to use this indirect way to subsidize banks rather than to give them the subsidy directly. However, it is a strange program indeed where banks get subsidized in proportion to how many "bad" assets they hold. This will make banks wish that they had made even greater mistakes, and held more assets that are likely to be truly worthless. However, these worthless assets could be worth a fortune to banks.
The Federal Reserve's unsound monetary policy in the early 2000s pushed down interest rates excessively, resulting in asset-price inflation, particularly in houses because they are bought primarily with debt. Eventually the bubble burst and house prices fell precipitately; they are still falling.
Becker's interesting post argues that the boom and bust in housing have not had as large an effect on consumption (and hence, this implies, on the nonfinancial economy) as the size of the price fluctuations might suggest. He illustrates with the example of a homeowner who has a bequest motive: if people intend to leave their house to their kids, changes in the value of the house will affect the size of the bequest rather than current spending by the owner-parent. More generally, an increase in home values increases the cost of housing by the same amount. If all house prices double (and assume no other prices change), but the owner is not intending to downsize, he cannot "spend" the increased value of the house.
However, although increased home values are unlikely to be translated into equal increases in consumption spending, those increased values are likely to have a strongly positive effect on consumption. To begin with, a significant amount of borrowing during the bubble involved the refinancing of existing mortgages rather than the financing of home purchases, and often the incentive for the refinancing was to obtain cash for consumption. Furthermore, some people may downsize, or even become renters, because they want to increase their consumption expenditures, as they can do if they cash out some of the increased market value of their house. And if people feel wealthier because the market value of their savings (which includes the value of a house) is rising, they may reallocate more of their income to consumption, which they can do without impairing the expected value of their savings if their houses are worth more. In other words, a house is a "store of value" (as economists say) rather than just a home, and even if one has no intention of downsizing, the fact that one has a valuable house that could be sold if one needed cash provides a store of value that may persuade you that you can afford to consume more. As a form of savings, a house is illiquid and risky relative to cash, but it is still savings, in the sense of an asset that one could turn into cash if necessary, to increase consumption in the future. And if one's savings shoot up because of an increase in the price of one of one's assets, one may decide to allocate a portion of the increased savings to current consumption.
These considerations persuade me that the run-up in housing prices probably did increase consumption significantly. More important, the collapse of those prices, together with the fall in the stock market, has almost certainly had a significant negative impact on current consumption expenditures. Because of rising house and stock prices, the market value of people's savings rose during the housing and stock bubbles and as a result people reduced their savings rate, to the point where it actually was negative for a period during the early 2000s and was no higher than about 1 percent before the crash last September. (This is related to the "store of value" point.) The personal savings rate has since risen to more than 4 percent. The fall in house and stock prices, combined with increased unemployment and fear of unemployment, convinced many people that they didn't have enough precautionary (safe) savings, and so their current savings are heavily weighted toward cash and other riskless, or very low-risk, forms of savings. The reallocation of income from consumption expenditures to very safe forms of savings reduces current consumption without increasing productive investment significantly, and so contributes to the depression.
One good aspect of the Treasury's plan to enlist the private sector in buying mortgage-backed and other bank assets is that it reduces the uncertainty-if it is implemented! - about what the government plans to do further in aiding banks. Starting with the vacillations of Henry Paulson, the former Treasury Secretary, the federal government's efforts to help banks have lacked a clear direction, and have wasted a lot of taxpayers money. Especially during a serious recession (I will call this a recession, not a depression, until the cumulative fall in GDP equals or exceeds 8-10 percent-so far the fall in US GDP has been about 2%, and world GDP has hardly fallen), consumers and businesses can cope much better if they know what the government plans to do. They can adjust much more easily to known government policies, even if they are not good policies, than to changing policies that lack any direction.
A major criticism of early plans for the government to buy bank assets through an asset auction was that the government would overpay for the assets since they did not know the worth of the assets offered to them. Although that difficulty might be overcome, the Geithner proposal uses government money to encourage hedge funds, pension plans, and other financial institutions to buy bank assets in order to use private competition to determine the worth of these assets. Hedge funds and other financial institutions do not want to overbid since that would reduce their profits from any future appreciation in the value of the assets bought. Competition among different financial intermediaries for these assets would prevent them from underbidding since they would then not be able to buy the assets.
To encourage private participation, the Treasury Secretary is offering bidders very generous terms. If say a hedge fund bids $100 for an asset, the fund would have to risk only about 7%, or $7. Another 7% would be risked by the Treasury (i.e., from taxpayers), and the rest would be a loan guaranteed by the Federal Deposit Insurance Corp. (FDIC). If the asset rises in value over time, the Treasury and the hedge fund would share the profits equally, while the hedge fund's losses if the price goes below $100 is limited to the $7 it puts up, no matter how low the price goes.
Therefore, the downside loss to private companies in this example would be sharply limited by the equity they put in, while the upside gain could far exceed their initial equity. This means that hedge funds and other funds would find riskier assets very attractive, and they would bid more for them than for less risky assets with the same expected return. For example, suppose one asset had a 100% chance of being worth $100 in the future. The expected value of the asset is obviously $100, but a private fund would bid $107 because the Treasury would pay $7 of this bid.
Suppose, on the other hand, there is another asset that has a 10% chance of appreciating to $1000, but it has a 90% chance of becoming worthless. The expected value of this asset is also $100, like the safe asset, but in my example it is worth much more to bidders under the Treasury's terms since the FDIC would pay the successful bidder 86% of its bid price if the asset became worthless. It can be directly shown that private funds bidding their expected value would then bid about $242 for this asset, which far exceeds the asset's overall expected value of $100 because the FDIC is guaranteeing most of the loss, and the fund would collect half the appreciation.
Even if it were desirable to subsidize private funds to bid for bank assets, is it wise to structure the subsidy in this way so that the bidding is skewed toward more risky assets? One reason for doing so is that assets with greater variability in their future worth are presumably harder to value. Hence banks holding these assets might value them more than other financial institutions would. These would then be the type of assets that banks would be reluctant to sell in an unsubsidized market since market bids would be below bank estimates of their value. The Treasury's approach raises the willingness to pay by hedge funds and other financial institutions for precisely such risky assets.
Posner's proposal is to do more of what the government did earlier; namely, lend to banks in return for preferred stock in the borrowing banks. This has the advantage of being simpler than the Treasury's convoluted proposal, and Posner gives some other advantages. However, I would worry a lot that the government when they hold greater amounts of stock would try to micromanage banks even in greater detail than they are already doing. Congress and the president have complained loudly about bonuses, pay levels, golf outings, and other business activities, and legislation was introduced to limit pay and perquisites. Under Geithner's plan, Congress might have less incentive to micromanage the decisions of hedge funds and others who buy bank assets since the government would have an equity interest in particular assets rather than an equity interest in the overall profits of these funds.
However, Congress would also complain a lot if hedge funds and others made a large profit from the assets they bought with government guarantees. Perhaps this is why the Treasury's proposal gives such a huge subsidy to the funds that would bid for bank assets. In the absence of large subsidies, leaders of these funds would be reluctant to expose themselves to the torrent of criticism and interference from Congress and perhaps also the President. Nevertheless, it is highly worrisome that taxpayers would become committed to such potentially large additional subsidies to the financial sector.
The Internet has produced revolutions in many kinds of behavior, such as greatly reducing the advantages from reading newspapers, but one of the most important is in providing easy access to various kinds of social interactions. The remarkable growth of social networking sites, such as Facebook and Twitter, reflect the desire of many individuals to communicate with other persons they do not know. The explicit goal of various networking sites, like E Harmony and Monster.com, is to solve age-old matching problems.
A challenge over many centuries is how to help eligible young men and women find suitable marital mates? Traditional societies use marriage brokers and relatives to bring young persons and their parents together, while in modern societies young persons meet through co-educational colleges, jobs, and dating. Both traditional and modern societies have struggled with the difficulty of finding suitable matches- Mrs. Bennett's lamented in Pride and Prejudice (published at end of 18th century) the difficulties of finding suitable husbands for her five daughters.
Similar matching problems occur in modern labor markets, where workers search for good jobs, and employers search for good workers. Search theory in economics shows how both workers and firms generally stop short of finding the best possible matches because of the cost in time, money, and passing up opportunities of continuing the search process. As a result, workers of given skills may end up with jobs that pay differently, and that differ also in other characteristics, while firms that pay similar wages may employ workers of different productivities.
The Internet provides significant opportunities to improve the search process, whether for men and women looking for suitable mates, or for workers and firms looking for good employment matches. It provides an especially good method of searching for individuals who are new to a city or other area, for persons who do not meet many potential matches through their work, schools, friends, or relatives, for those who want to cut down on the time spent searching, and for others who do not want their friends or others to know that they are actively searching. According to its publicity, eHarmony, an important online marital matching site, "matches you based on compatibility in the most important areas of life - like values, character, intellect, sense of humor, and 25 other dimensions."
Online matchmaking has grown rapidly during the past decade. Some estimates claim American online dating and matchmaking companies take in over $1 billion per year. Estimates for China and India, probably exaggerated, are at $1.3 billion and $300 million, respectively. Revenues from online dating sites continue to grow rapidly, especially in Asia, although the worldwide recession has slowed down these growth rates. These data exclude men and women who meet online not through official matching sites, but through more informal social networking, as in chat rooms and Facebook.
It is difficult to know precisely how many men and women marry because they met their mates through online matchmaking sites. I know several friends who did, but perhaps I know busy or hard to please men. Match.com says it has more than 15 million members. A study by Harris Interactive conducted for eHarmony finds that over 86,000 eHarmony users married between April 2006 and March 2007, which is three times the number estimated for 2005 in a different study. The Harris study also estimated that over two percent of all persons marrying between April 2006 and March 2007 were introduced through eHarmony.
Economic theory and empirical evidence (see my A Treatise on the Family) indicate that men and women who marry tend to have similar backgrounds by religion, race, education, family income, and many other characteristics. This is called positive assortative mating. Since marriages between persons of similar characteristics usually (but not always!) produce more satisfactory relationships, it is not surprising that marriages between unlikes; that is, persons with quite different characteristics, such as by education, religion, or race, are far more likely to end in divorce than are marriages between similar persons.
I would expect matches made online,just like those made by traditional matchmakers, to generally bring together persons with rather similar characteristics since the preferences of individuals, and also the matching algorithm used by different online sites, would tend to match individuals with similar backgrounds. Given the analysis and evidence cited in the previous paragraph, this would imply that marriages brought about online would tend to be more stable than other marriages. I have not seen any studies that try to test these theoretical implications on the degree of assortative mating and the marital stability of online matches.
The online job search market is much larger than that for dating and marital matching. Help wanted ads used to be an important source of income for local and even larger newspapers, but most of that market has dried up because job search shifted online. This has been one of the important factors in the sharp decline of newspaper revenues, and the despair gripping that industry.
Online employee-employer matching companies claim many advantages for their method of matching workers to jobs. Foe example, the company OpenHire asserts in its advertising that it will "Centralize your recruiting into one system that reduces data entry, eliminates processing service fees and controls agency fees. OpenHire integrates with e-mail, online job boards, corporate job sites and internal intranets..."
By lowering search costs, online matchmaking for the marriage and labor markets, and for other markets as well, improve the efficiency of these markets by reducing the variance in the quality of the matches to persons with given characteristics, whatever the characteristics are. In the future these markets will have more of an international flavor, where individuals will search online for good job or marital matches in countries other than where they are living. Online matchmaking, whether formal or informal, will increasingly have major implications for the sorting of individuals in marriages, jobs, and in other types of matches, implicatiosn that would have been extremely difficult to foresee in the early days of the Internet.
Matching is a form of search in which the object is to create a relationship, such as marriage or employment (marriage could be regarded as a form of employment). Matching is more complex and often more protracted than most searching for goods or services because the stakes tend to be greater. The costs of exit may be high and often there are high opportunity costs as well. The higher these costs of mistaken matching, the more it pays to invest in the search for a good match. But a protracted search is very costly; for example, if one spends 10 years searching for the perfect mate, one has lost 10 years of benefits of marriage.
If one divides the process of searching for a match into two parts, which I'll call "screening" and "meeting," one can see more easily the precise benefits of Internet matching services. The universes of potential mates, workers, and jobs are immense, and while very few members of these universes are suitable candidates for a particular match, it is very difficult to determine in advance who those few might be. So the first task in the search for a match is to screen out the vast number of unsuitable candidates, and this is done much more quickly, completely, and accurately (and therefore costs less and confers greater benefits) by an Internet search than by such traditional alternatives as mixer dances, marriage brokers, the personal columns in magazines, wanted ads, and job fairs. Once the candidates for a match have been reduced to a manageable number, the more time-intensive, face-to-face meeting phase of the matching process takes over; an efficient preliminary screening greatly reduces the aggregate costs of the expensive "meeting" phase of matching. In job search, the meeting phase is the interviewing of the most promising candidates.
However, two forms of non-Internet marriage-matching screening should be mentioned, as they are also relatively new (and increasingly common) and highly efficient. They are coeducational higher education and gender integration of the workplace. In effect, both the college, which wants a homogeneous student body, and the employer, who wants a homogeneous work force within each job category, do the prescreening at no cost to the searcher, and the fact of taking classes with, engaging in extracurricular activities with, or working side by side with someone of the other sex reduces the incremental costs of the "meeting" phase of the search: you don't have to go out of your way to meet someone you work with or go to class with.
My guess would be, therefore, that the demand for Internet marriage screening would be less among students and among young workers in workplaces that have a fairly even balance of men and women, and greater among persons who are not in such advantageous situations and among persons who have high opportunity costs of time, like successful businessmen and professionals. It should also be greater among persons who have idiosyncratic or minority tastes. For example, since homosexuals are a relatively small fraction of the population, I would expect their demand for Internet match-screening services to be proportionately greater than that of heterosexuals. Personal ads in magazines that have a specialized readership are another form of non-Internet preliminary screening.
The advantages of Internet job matching over newspaper want ads are particularly great, and this for four reasons. First, in any community in which there is more than one newspaper, the job searcher (whether looking to be hired or to hire) has to buy and read both newspapers (or however many there are), even though he may derive no value whatsoever from anything in the second through nth papers except the want ads. Second, a job search is often regional or national rather than local, and it is infeasible to do a regional or national job search by means of local newspapers. Third, the costs of paper greatly limit the number of jobs that can be advertised in a newspaper and the amount of information that can be conveyed about each job or job hunter. And fourth (though related to the first point), a newspaper is a bundled commodity, and people searching for jobs or workers may derive very little value from the other sticks in the bundle.
Of some $300 billion in annual American charitable giving, about 5 percent is spent abroad; almost 40 percent of that 5 percent is donated by the Bill & Melinda Gates Foundation, mainly for trying to alleviate Third World health problems (such as malaria and AIDS) and provide assistance to Third World agriculture. Total American charitable giving abroad is more than half as great as total U.S. governmental foreign aid, of which almost a third goes to Israel and Egypt.
Americans do not receive an income tax deduction for giving to foreign charities, but they do for giving to domestic charities that donate abroad (provided they don't just donate to foreign charities). Should they? I am inclined to think they should not.
This is not, however, because I think that government foreign aid is a more efficient method of increasing Third World welfare. Little U.S. foreign aid goes to the Third World, and the efficacy of the aid that does go there is undermined by the requirement that the aid be used to purchase U.S. goods and services.
Nor do I question the economic rationale for providing a tax exemption for charitable donations. The argument is that charitable giving provides an external benefit; that is, if I want to increase the welfare of people in Bangladesh, I will benefit from a charitable donation for that purpose made by someone else and will therefore be inclined to give less. Knowing this, that donor will donate less because the value of his donation is diminished by my free riding on it. There is even an argument that placing any restriction on how a person uses his money reduces incentives to earn, but it is hard to believe that this would be a big effect of eliminating the tax exemption for charitable donations to foreign countries.
Although charitable donations to foreign recipients will thus diminish if the tax deduction is repealed, there is likely to be some substitution in favor of domestic recipients, which will increase welfare in the United States; and that seems to me a good thing, especially in a depression. (Of course, the depression may induce donors to reallocate grants to the United States because the wealth of Americans is less; but perhaps not, because the decline in wealth in the Third World is probably as great or greater.) Also, to the extent that total charitable donations fall, tax revenues will rise, which is also a good thing, given the enormous budget deficits that we face. I suspect, moreover, that charitable donations to Americans create more utility than charitable donations to people in poor countries, because the latter donations reduce pressures for desperately needed political, economic, and social reforms. I doubt that Mugabe would still be ruling Zimbabwe if the West did not provide extensive food to its starving population. I have not seen any attempt at a rigorous analysis of the net benefits of charitable contributions to Third World nations.
There is also some danger that charitable donations to foreign countries undermine U.S. foreign policy. But I do not put much weight on this factor, because the government can forbid donations to countries with which we are at odds, or to other countries where we think aid would undermine our foreign or security objectives. An offset, moreover, is that donations from the United States, even if made by private individuals and foundations rather than by the government, build good will toward Americans.
Pressures for both increased government and private assistance to poorer nations may well result from this recession because some nations that can least afford a setback may suffer the most due to reduced demand for their exports of commodities and other goods. In considering the topic of private charitable giving to other countries, Posner stresses one of the economic rationales for tax exemptions of charitable giving; namely, that tax deductions are a recognition of the interdependence among individuals in the benefits from charitable activities, such as those by hospitals or schools. If many individuals and organizations benefit when recipients of their charity has more resources, each donor will tend to under give because they do not take account of the benefit to other donors from their donations. That is, giving by any individual or organization produces a positive or beneficial "externality" because it adds to the welfare of other givers.
Another important rationale for tax exemption of charitable contributions is to decentralize giving away from the government and toward the private sector. Just as government grants to hospitals and other beneficiaries crowd out giving to these beneficiaries by private foundations and religious organizations, so too does private giving reduce the need for government giving. For example, when private universities obtain support from wealthy individuals and foundations, this allows them to compete more strongly against public universities, and thus reduces the need for as much public funding of higher education. This "crowding" out of government giving may be valuable because private donations are generally better thought out and more efficient than public grants and aid.
Both arguments seem to apply rather fully to giving to foreign charities as well as to domestic ones. Private giving to help foreign hospitals, doctors, schools, or individuals reduces the need for foreign aid by the US government to the same type of organizations and individuals. This is preferable because government foreign aid invariably goes through other governments, and hence tends to be centrally controlled, and subject to government inefficiencies and corruption. Private giving is more effective at getting the assistance to those who need it.
Giving by an American individual or organization to foreign charities also creates an externality to other givers and to others who benefit as well from such giving. There is a positive externality even if one counts only the benefits produced to other Americans, as long as many American individuals and organizations benefit from the giving to particular foreign charities.
That both arguments for tax deductions are applicable to foreign giving makes a case for also allowing giving to foreign charities to be tax deductible. There is, however, one important argument against doing so: the difficulty of determining whether foreign recipient charities are legitimate and acceptable charities. If such giving were deductible, American individuals or businesses might try to funnel assets into fake foreign "charities" that they control in order to reduce their taxes. Such tax evasion may be relatively easy for American authorities to determine for American charities, but extremely hard for them to pinpoint in poorer countries with limited data and few investigations of fraud. It may also be difficult to determine if foreign recipient organizations are spending the contributions received from Americans on desirable purposes rather than in promoting terrorism and other activities detrimental to US interests.
Still, I tend to support allowing at least some tax deduction for giving to foreign charities. Otherwise, it may be just another form of protectionism, where American services and goods are favored over foreign services and goods. Since protectionist arguments take many disguises, it is likely that some of the opposition to allowing tax deductions for foreign charities is due to the desire to impose tax disadvantages on the "import" by American individuals and organizations of foreign "goods".
The severity of this recession has stimulated calls for greatly increased regulation of the financial sector, and for changes in some of the present regulations. Some new regulations are desirable, but the type of new regulations must be in response to a recognition that regulators failed in this crisis because they did not use the authority they had to rein in some of the investor exuberance.
The claim that the crisis was due to an insufficient level of regulation is not convincing. For example, commercial banks have been more regulated than most other financial institutions, yet commercial banks performed no better than other classes of financial institutions. At the other extreme, hedge funds have been the least regulated, and on the whole they did better than most others in the financial sector. One major problem with regulations is the regulators themselves. They get caught up in the same bubble mentality as private investors and consumers. For this and other reasons, they fail to use the regulatory authority available to them. This implies that as much as possible, new regulations should more or less operate automatically rather than requiring discretionary decisions by regulators.
Many critics have blamed part of the financial crisis on the requirement that financial institutions value their assets at market prices rather than historical costs. One problem with such "mark to market" pricing that became apparent during the crisis is that it is difficult to accurately value assets in very thin markets that have few transactions. Moreover, most of the transactions that do take places are fire sales made because sellers need quick cash. In such markets, if assets are held for a while before they are sold, their value may be much higher.
One does need flexibility in using the mark to market principle, but this accounting method is most of the time a much more useful way of valuing assets than using original cost of the asset, even when adjusted for depreciation. For assets that have been held for significant periods of time may be subject to huge changes in worth that make original cost largely irrelevant.
Perhaps a useful approach (suggested to me by David Malpass) is that instead of requiring all companies to use either mark to market or cost based accounting, companies should be permitted to decide which method to use. This would add a little to the information complexity of evaluating company assets, but it would also make the accounting process more flexible. Presumably, however, companies should have to commit to one or the other accounting rules for some timee rather than being allowed to change their approach whenever they see fit.
Once we are out of this crisis but not before we are out, I believe capital requirements should be imposed on investment banks, hedge funds, and other financial institutions in the form of maximum allowable ratios of assets to capital. One major advantage of such a requirement is that it can operate rather automatically rather than requiring regulators to make discretionary choices. The extremely high leverage in many financial institutions during the past few years created a fundamental instability in the financial sector regarding its ability to respond to large negative aggregate shocks to the system rather than only individual firm idiosyncratic shocks. Limiting the ratio of assets to capital would help prevent the high leverages that contributed to the collapse of many financial institutions in the wake of the sharp falls in the values of the assets they were holding.
Capital requirements also provide a way to respond to the "too big to fail" principle when, rightly or wrongly, large firms are often kept from going bankrupt. When large financial firms get into trouble, they impose costs on everyone else both due to the repercussions on financial and other markets, and to the taxpayer monies used to bail them out to prevent their complete collapse. For example, during this crisis the sharp declines in the values of the diversified commercial bank Citigroup, and of AIG, a giant insurance company and in recent years hedge fund, imposed major costs on the system. Their collapse led to massive and continuing federal government injection of monies into these companies.
One-way to reduce the likelihood of a too-big-to fail-problem is to impose higher capital requirements relative to assets on larger financial firms. That is, to implement a progressive set of capital requirements relative to assets that would increase as the size of a bank or other financial firm increased. Since they would not be allowed to expand so much beyond their capital base, larger financial institutions would be better prepared to deal with aggregate shocks to the financial system than they were during this crisis.
Such a progressive system of capital requirements would also reduce the incentives to become large since this system would impose a "tax" on becoming big. In an environment when large firms are protected by the government from failing, and when their failure helps bring down other interconnected financial and other firms, decreased incentives to become a large financial institution are desirable because of the cost these institutions impose on everyone else.
Beginning in the 1970s, the banking industry was extensively deregulated. Other financial intermediaries, such as broker-dealers, hedge funds, and money-market funds, were permitted to offer close substitutes for services provided by commercial banks, and restrictions on banking were loosened so that the banks could fight back against their new competitors. The deregulation program was complete by 1999, when the Glass-Steagall Act, separating investment and commercial banking, was repealed. However, the Bush Administration, as part of its general free-market philosophy, instituted a regime of regulatory laxity that included bank and securities regulation. This laxity, along with the Federal Reserve's error in depressing interest rates in the early 2000s, contributed to precipitating the banking collapse and the ensuing depression in which we find ourselves.
A natural response is to tighten up regulation. In the case of commercial banks, this would not require new legislation. The bank regulators have virtually plenary control over banks: thus the crack "what does a bank say when a regulator tells it to jump?" Answer: "How high"? Burned by the banking collapse and employed by an Administration less complacent about the self-correcting character of financial competition than the Bush Administration, current regulators will not allow banks to take risks though, paradoxically, may compel them to in order to increase the amount of money in circulation and thus stimulate economic activity. The banks, being undercapitalized, are afraid to make risky loans; they thus don't have to be prevented from taking them, at least in the near term. The current complaint about the banks is that they are hoarding cash--that they are excessively risk averse--and thus are failing to provide the credit that the economy needs in order to recover. To tighten regulation of banks at this point would thus not only be a case of closing the barn door after the horses have escaped, but also would undermine the government's policy of encouraging banks to lend.
The most challenging issue of financial re-regulation is bringing the nonbank financial intermediaries under the regulatory umbrella, in order to prevent effective bank regulation from simply shifting ever more financial intermediation to firms not shackled by regulation. One can imagine imposing capital requirements, leverage limitations, or even reserve requirements, on nonbank financial intermediaries. But this would require an elaborate regulatory apparatus that would cost a lot and, more important, might be ineffectual because of the complexity of modern finance and the heterogeneity of the nonbank intermediaries. I would prefer to see, at least as an initial step, requiring greater regulation of specific financial instruments, in particular credit-default swaps, which are at present unregulated credit-insurance undertakings often with no backing in the form of either reserves or collateral. Financial intermediaries find themselves both issuers and purchasers of such swaps (that is, both insurers and insureds), and because the swaps are not traded on an exchange, are not standardized, and are not regulated to assure that the issuer can honor his undertaking, they have been a source of debilitating uncertainty in the present crisis.
We would not be in the fix we're in were it not for the Federal Reserve's having pushed down interest rates too far and keeping them down too long, thereby setting the stage for a credit binge (including the housing bubble), and, relatedly, were it not for the very low personal savings rate of Americans and their investing almost their savings in risky assets such as houses and common stocks. The problem of excessive borrowing can be addressed both by the Federal Reserve, which exercises a high degree of control over interest rates, and by the government's placing limits on credit-card and mortgage debt, for example by repealing the deductibility of mortgage interest from taxable income.
But the most important point I would make is that there should be no new regulatory measures until the depression reaches bottom and recovery begins (not that there can be certainty about when that point has been reached--there were several false bottoms in the 1930s depression). Any regulatory initiatives at this time will simply increase the already great uncertainty in which the financial industry is operating; and as Keynes pointed out, anything that increases uncertainty in a depression causes hoarding, which can in turn precipitate a deflation likely to deepen and protract an economic downturn.
