Trade policy

The Clinton Administration’s position on trade has been virtually identical to that of its Republican predecessors, proclaiming the universal virtues of free trade and pushing for Presidential authority to negotiate so-called ‘fast track’ agreements to further it, by-passing normal legislative scrutiny. Gestures towards labour or environmental concerns have been almost completely empty of content—the sound-bite sop to demonstrators at Seattle is a good example.footnote⁴ The regime’s actual position on trade is set out at length in the Economic Report of the President (ERP) for 1998, which under the rubric of ‘Benefits of Market Opening’, rehearses the standard neo-classical case for free trade—i.e. the Hecksher–Ohlin argument for efficiency gains through specialization, especially as economies of scale enhance productivity. Unsurprisingly, longstanding critiques of the assumptions behind the Hecksher–Ohlin model—in particular, its premisses of full employment and comparable technologies among trading partners—are ignored. Without these assumptions, it does not follow even from the orthodox model itself that all countries will necessarily benefit from trade opening, since liberalization can, for example, trigger a rise in unemployment.footnote⁵ However, even if one accepts the assumptions, it is still well-known that trade opening produces losers as well as gainers. The Hecksher–Ohlin model itself stipulates a tendency toward factor price equalization among new commercial partners, which implies that when trade opens between a high and low wage country (for example, the US and Mexico), downward wage pressure will be felt among the workers in the high-wage economy. Within this framework, it has therefore long been understood that for trade to be equitable as well as efficient, even by minimal Pareto standards, losers from liberalization need to be recognized and compensated for their losses.

If it is difficult to estimate the precise impact of NAFTA, or any other single trade agreement, on the work and wages of average workers in the US, another question can be dependably answered. Rhetorical flourishes aside, the Clinton Administration has offered virtually no compensation to workers whose incomes and/or job security have been hurt by its trade policies.footnote⁶ Rather, the 1998 ERP reiterates mainstream arguments that trade is not primarily responsible for either the long-run wage stagnation for most American workers or the increased differentials between high- and low-wage workers. Rising inequality in the labour-force it attributes, instead, to ‘skill-biased technological change’, as the introduction of new, computer-based processes creates wage premia for workers able to handle them while depressing the income of workers unable to do so. David Howell has demonstrated fundamental flaws in this view, showing that increased wage inequality is actually more a reflection of social and institutional, rather than technological, changes—in particular, the steady weakening of American trade unions and growing hostility of American labour laws to the concerns of working people.footnote⁷ It follows that even if the Clinton Administration has failed to address losses to US workers from its trade policies, it could still have compensated them with labour market policies and measures to redistribute income. What has been its record in these areas?

Labour policy

The short answer is that the Clinton Administration has done virtually nothing to advance the interests of organized labour or working people more generally. As longtime labour journalist David Moberg has commented, ‘Clinton has probably identified less with organized labour than any Democratic President this century.’footnote⁸ Of course, since the AFL–CIO is a permanent electoral prop of any Democratic candidate to the Presidency, its concerns cannot be completely disregarded in the Republican manner. Clinton thus supported a two-step rise in the minimum wage in 1996–97, from $4.25 to its current level of $5.15. But this modest increment has done little to reverse the precipitous fall in the real value of the minimum wage. In 1996 the real value of the $4.25 rate was more than 40 percent below its buying-power in 1968. At the new rate of $5.15, set in September 1997, the minimum wage is still over 30 percent below its real value in 1968, even though the economy has become 50 percent more productive over the past thirty years.footnote⁹

The Administration also claims the Family and Medical Leave Act of August 1993 as a major accomplishment. The law requires employers with 50 or more employees within a 75-mile area to offer 12 weeks of unpaid leave per year for employees who had worked at least 1,250 hours within the past year, to be taken for health problems, the birth or adoption of a child or care for a family member. The exact boundaries of these grounds have been poorly defined, as also the conditions that employers—who may require that employees use up their sick and vacation days before taking advantage of the Act—may impose on them (can the leave be taken in blocks of a few hours, creating in effect a part-time job with full benefits, etc?). In the area of worker training, billed as a central plank of Clinton’s pledge to ‘Put People First’, the Workforce Investment Act of August 1998 has consolidated the 40-odd Federal training programmes, introduced vouchers for workers to pay for private training, and created a nationwide jobs data-base.

Proposed in March 1993, but not passed, was a Striker Replacement Act that would have barred companies from permanently replacing strikers in disputes over working conditions (temporary replacements and strikes over pay were not to be affected), which eventually got fewer votes in the Senate than a similar bill under Bush. However, in perhaps the best example of the President’s gesture politics, in March 1995 Clinton signed an executive order barring Federal contracts of over $100,000 to companies which had permanently replaced striking workers. Initially the Senate threatened to block the directive, but in the end it desisted, perhaps persuaded by reported Administration arguments that virtually no major contractors would be affected. In February 1996 the order was struck down by the DC Circuit Court of Appeals, and the issue was dropped. During the year it was nominally in effect, not one contract was cancelled under the directive. On the other side of the ledger, Clinton did veto cuts in funding for the National Labor Relations Board and Occupational Safety and Health Administration, and a bill that would have effectively legalized company unions; and his appointments to the National Labor Relations Board, in particular Frederick Feinstein as general counsel, have been more favourable to trade-union leaderships than before. But the overall record of the Administration on labour issues is in any comparative sense remarkably thin, especially given its unqualified support for free trade in face of labour opposition. The net impact of Clinton’s tenure can be seen from the fate of organized labour itself. Far from picking up after its long decline, union membership fell even further during the Clinton Presidency. In 1998 it stood at 13.9 percent of the total work-force, nearly three percentage points below the 16.8 percent to which it had dropped in 1988, Ronald Reagan’s last year in office.

Fiscal Policy

Clinton’s tax policies, lessening the highly regressive impact of the Reagan–Bush years, constitute the principal item in the Administration’s claims to represent an enlightened alternative to the Republican rule of the eighties. The omnibus Budget Reconciliation Act of 1993, which raised taxes by $240.4 billion over 5 years, increased the levy on incomes over $140,000 from 30 to 36 percent, with an additional 10 percent surcharge for incomes over $250,000. It also included a higher gasoline tax, a greater cap on income subject to Medicare hospital insurance tax, and a substantial extension of the Earned Income Tax Credit (EITC), or direct income supplements to low-wage earners. Clinton’s second major piece of legislation was the Taxpayer Relief Act of 1997, which reduced taxes by $290 billion over 10 years—that is, reducing by half the revenue gains scheduled in the 1993 Act. The new law combined a range of child and education tax credits with lower taxes on inheritance and capital gains. Calculations by the Citizens for Tax Justice estimate that the net effect of the 1997 Act has been to cut taxes for the top 60 percent of the population, with the great bulk of the hand-out going to the top 20 percent.

Mishel, Bernstein and Schmitt have compared the total impact of the Clinton programme with the Reagan record by analysing the Federal tax rates of households at 1998 income levels, under the tax laws that prevailed in various years between 1977–98. Table 1 summarizes their main findings.

As the table shows, between 1977–89, the bottom 80 percent of households experienced a slight increase of 0.5 percent in their tax obligations between 1977–89, and a somewhat bigger decline in these of 1.6 percent between 1989–98. Most of the tax reductions were concentrated among the least well-off 20 percent of households, who experienced a 4.6 percent decline in their fiscal burden—a drop primarily due to changes in the Earned Income Tax Credit. Correspondingly, the richest 20 percent of households experienced a small decline in their tax rates of 1.3 percent between 1977–89, which was reversed between 1989–98 when their rates rose by 1.4 percent. Here, by far the biggest swing was among the top 1 percent, who experienced a 13.4 percent tax cut between 1977–89 and a 7.7 percent tax increase in the period 1989–98. Thus the Clinton Administration did restore part of the progressive dimension of the tax system that was lost in the 1980s, but not all of it.

What is not clear from the table is the extent to which the stock market boom, and consequent rise in revenues from Capital Gains Tax, helped to swell government coffers and allow a balanced budget by 1997. Between 1992–97, revenues from capital gains rose from $126.7 to $362 billion—jumping from 2.7 to 5.0 percent of Treasury receipts. This is just one indicator of the extent to which Clinton’s economic record has depended on the fortunes of Wall Street, a topic we consider in more detail below.

Expenditure policy

The over-riding objective of the Clinton Administration has been to bring government expenditures down, in line with its broader macroeconomic priority of deficit reduction. The extent of the spending cuts is set out in Table 2, which shows Federal Government expenditure patterns from 1992, the last year of the Bush Administration, until 1999. Between 1992–99, total Federal expenditures fell as a percentage of GDP from 21.9 to 18.6 percent, a decline of 14.9 percent. The most significant reduction slashed military spending from 4.7 to 3.0 percent of GDP between 1992–99, in absolute terms a drop of 36.7 percent. But there have been large cuts in other areas as well, including support for education (-9.2 percent), science (-19.1 percent), transportation (-11.2 percent) and income security (-16.0 percent).

The cuts in military spending, of course, reflect the end of the Cold War and consequent expectation of a widespread ‘peace dividend’. While they have been substantial, what is more remarkable is that the annual military budget should have remained at $300 billion, after the justification for the exorbitant arms race of the Cold War period has evaporated. Spending on arms remains 4.6 times greater than Federal outlays on education. It is also triple the size of the military budget of the 1930s as a proportion of GDP—the last decade before the World War and Cold War, when GDP itself was historically low. The fact is that in so far as the end of the Cold War has yielded a peace dividend, it has taken the form of an overall decrease in the size of the Federal Government rather than an increase in Federal support for any of the programmes supposedly cherished by Clinton, such as better education, improved worker training, or alleviation of poverty.

The Administration’s extension of the Earned Income Tax Credit has been its most significant anti-poverty initiative. The EITC originated under Ford’s Republican Administration in 1975, when it covered 6.2 million families for an average credit (in 1998 dollars) of $609 per family. The programme expanded under Carter, Reagan and Bush alike (1978, 1984, 1986, 1990). By 1992, the last year of the Bush Administration, it covered 14.1 million families, who received an average credit of $1,076 (in 1998 dollars). By 1997 Clinton’s add-on had extended it to an estimated 19.5 million families for an average credit of nearly $1,600. Against this enlargement must be set the dismantling of welfare assistance programmes—what had been called Aid to Families with Dependent Children (AFDC), and is now termed Temporary Assistance for Needy Families (TANF). Thus, while outlays for EITC rose (in 1998 dollars) by $14.7 billion, from $8.5 to $23.2 billion between 1992–98, spending on ‘family support’ fell by $2 billion, from $17.5 to $15.5 billion. This was not the only area of welfare to contract under Clinton. Expenditure on food stamps and other nutritional assistance dropped by $4.3 billion, from $37.8 to $33.5 billion between 1992–98—a decline reflecting an increase in the percentage of households who are not receiving food assistance even though their income level is low enough to qualify them to receive it. Between 1995–97, the decline in the number of people receiving food stamps—4.4 million—was five times greater than the decline in the number of people living in poverty, and was accompanied by a dramatic increase in the pressure on private soup kitchens and food pantries.

While the full reasons for the fall in public food assistance under Clinton are not entirely clear, it appears that the primary factor has been his campaign to ‘end welfare as we know it’. The official attack on dependency has led both to greater stigma in receiving public assistance, and to greater difficulties in securing food stamps. When the pre-Clinton welfare system was still functioning, a high proportion of recipients took their food stamps and cash assistance at the same time.footnote¹⁰ The result of Clinton’s policies has been a complete standstill of expenditure on all public and food assistance programmes as a proportion of GDP, which remained at 1.3 percent in both 1992 and 1998. Clinton’s defenders, of course, would reply that the Administration has replaced a bad programme—welfare—with a good one, because the EITC creates incentives for work and does not discourage parents from living as a family unit. But if the EITC does correct some of the failings of the older welfare system, it also creates new and no less serious ones. Moving poor and unskilled women from welfare onto the labour market exerts downward pressure on wages, and the minimum wage itself is still too low to allow even a full-time worker to keep just herself and only one child above the official poverty line. A woman in this position will receive EITC to supplement her income, but she will also now have to pay for childcare and get less support from food assistance programmes. Structurally, the interactive effects of a low minimum wage and widened EITC have allowed business to continue to offer rock-bottom wages, while shifting onto taxpayers the cost of alleviating the poverty of even those holding full-time jobs. In sum, the Clinton Administration’s anti-poverty programmes have not increased income transfers to the poor as a share of GDP. After allowing for the increased costs of childcare, the overall conditions of life for America’s most destitute households may have worsened during the Clinton Administration.

Macroeconomic and regulatory policies

Throughout, the over-riding economic objective of Clinton’s Presidency has been deficit reduction. At its outset, the Administration adopted the doctrine that stringent control of Federal expenditure was a condition of lowering interest rates, which alone could stimulate private-sector investment and therewith overall growth. We have seen the pattern of spending cuts to which this commitment has led. Even after a balanced budget was attained in 1997, Clinton pressed on—now with the aim of reducing and even eliminating outstanding Federal debt. Concomitantly, monetary policy has been loosened, as Greenspan at the Federal Reserve has tolerated lower rates of unemployment than Volcker, while remaining vigilant against upward wage pressures. Nevertheless, interest rates have remained well above historical levels.

Financial markets, meanwhile, have been been treated with a more generous hand than labour markets. The Federal Reserve and the Treasury organized lavish bail-outs for investors in the Mexican, East Asian and Long-Term Capital Management crises—interventions that succeeded in preventing any cascading collapse in private credit markets, and protecting the interests of major wealth-holders. By validating speculative operations that would otherwise have ended in disaster, these bail-outs have, of course, been primers in ‘moral hazard’, setting the stage for further excesses that could trigger more severe crises later on. But this liberality is consistent with its approach to financial regulation at large. For the other major landmark of the Clinton Administration has been repeal of the Glass–Steagall legacy of the Depression years. The barriers Glass–Steagall set up between commercial and investment banking, and its obstruction of banking across state lines, were in practice no longer effective instruments of financial stability, or of widespread access to credit, under conditions of global integration. But no attempt was made by the Clinton Administration to seek more contemporary means of securing the stabilizing and redistributive ends Glass–Steagall had aimed at, such as a combination of taxes on speculative transactions and lower reserve requirements on loans for productive investments.footnote¹¹ Quite the contrary. Emblematically, Treasury Secretary Robert Rubin, after orchestrating the elimination of Glass–Steagall, was among the first to benefit personally from it, moving from Washington to become Co-Chairman of the newly merged banking conglomerate Citigroup—a fusion of the former commercial bank Citicorp and the former investment house Travelers that would never have been permitted under any reasonable interpretation of Glass-Steagall. In short, by both deregulating financial markets and diminishing the supply of low-risk Treasury securities,footnote¹² the Clinton administration has done much to encourage speculative over-drive.

Economic performance in the United States under the Clinton Administration has been widely hailed as an unqualified success.footnote¹³ But looking at some basic indicators in a comparative historical perspective, presented in Tables 3–6 below, we observe a much more mixed performance. There have been some significant departures from historic trends. But, given the centre-right direction of policy under Clinton, it should not be surprising that their benefits have been skewed toward the wealthy. Moreover, virtually all the economic achievements of the Clinton years have been tied to the extraordinary performance of the stock market, which has concurrently generated a highly fragile financial structure.

Table 3 presents some basic macro statistics—GDP growth, productivity, unemployment and inflation. In the upper panel of the table, the data are grouped by presidential eras—I have combined Kennedy–Johnson, Nixon–Ford and Reagan–Bush, as well as showing the Carter and Clinton years separately. In the lower panel, I group the same data according to NBER business cycles, as a check on the reliability of presidential eras as a measure of economic trends.footnote¹⁴

These indicators make it clear that, at least through 1999, the Clinton years have not been unusually successful in historical terms. Most strikingly, the Clinton period has not approached the macro performance of the Kennedy/Johnson era, when both GDP (4.8 vs. 3.7 percent) and productivity growth (3.4 vs. 1.8 percent) increased much more rapidly, while average unemployment (4.8 vs. 5.6 percent) was substantially lower. The figures for Clinton will improve when 2000 is included in these aggregates. Nevertheless, even allowing for an additional strong year, the Clinton performance will not match that of the Kennedy/Johnson era in GDP growth, productivity or unemployment. On the other hand, the rate of inflation under Clinton has been kept down to the lowest range level of the Kennedy/Johnson years, and declined over time, whereas inflation took off towards the end of Johnson’s Presidency as costs of the war in Vietnam escalated. Of course, a decline in inflation in itself does not tell us much about who gains or loses from it—it might indicate slack labour markets of no benefit to wage-earners.

Judged by less stringent standards than the 1960s, however, the macro-economic record of the Clinton years compares favourably with that of Nixon/Ford, Carter and Reagan/Bush. GDP growth was higher and both unemployment and inflation were lower. Productivity growth was still slow, even relative to the Nixon/Ford years. But the overall performance of the American economy has been stronger, if not to a dramatic degree. Clinton’s tenure, of course, has been graced by the circumstance that no recession has occurred during his Presidency.footnote¹⁵ If we consider the relative performance of the economy in the full business cycle of 1991–98 as against its predecessors, the 1990s do not stand out relative to either the 1970s or the 1980–90 cycle, and do still worse relative to the 1960–69 cycle. In short, it is hard to make a serious case that the US economy in the 1990s has been unusually robust once we take account of the 1990–91 recession. Of course, supporters of Clinton would claim that his ability to avoid a recession since 1993 has been a major accomplishment in itself.

Changing composition of GDP

Further perspective on the macroeconomic record of the Clinton years is offered by Table 4, showing the breakdown of GDP into component expenditure categories—consumption, government, investment and net exports. Two sets of figures stand out here. The first we have already noted: there has been a drastic contraction of government spending, which at 18.2 percent of GDP is far below that of any previous presidential period. What we also see in Table 4 is that the slack created by the fall in public expenditure has been taken up by rocketing private consumption, which at 67 percent of GDP is more than five percentage points higher than during the Kennedy/Johnson boom. It is clear from these figures that it is the rise in consumer spending that has been the driving force of aggregate demand under Clinton, allowing government expenditure to fall without generating a slowdown in overall growth. Thus, to understand what has sustained growth in these years, we need to look at the bases for the expansion of private consumption. A consideration of financial market practices and performance in the Clinton period throws a good deal of light on this question.

Financial market behaviour

The most dramatic economic change of the Clinton presidency has been the transformation of the country’s financial structure by the stock bubble and shifts associated with it. Table 5 provides some indication of what has been involved. During the Kennedy/Johnson and Reagan/Bush periods, the Standard and Poor index of the stock prices of the top 500 companies in the economy rose at a rapid annual rate of 6.2 percent. During the Nixon/Ford and Carter years, the S&P actually fell in real terms. Under Clinton, it has registered an annual growth rate of 17.6 percent that has no historical precedent.

The performance of the stock market under Clinton becomes even more astonishing when measured against GDP during the various presidential eras. In theory, fluctuations in equity prices over a full business cycle are supposed to reflect the underlying performance of the real economy. Thus, by measuring the difference between growth of the S&P 500 and GDP, we can observe the extent to which the stock market is responding to real economic developments. Here again, the Clinton experience is without precedent. Since 1993 the rise in stock prices has been 13.9 percent above that of the real economy. Even in the Reagan and Bush years, during which economic policy overwhelmingly favoured the prerogatives of capital, and financial capital in particular, stock prices rose only 3.3 percent faster than GDP.

Table 5 also presents some data on changes in household financial patterns during the Clinton boom. The third row of figures suggests the degree to which the consumption boom has been debt financed. Household debt—including mortgage and consumer debt—has ratcheted upward dramatically during Clinton’s tenure, to reach 94.2 percent of disposable income. This compares with a ratio of 77.8 percent during the Reagan/Bush years, itself an unprecedented level compared with any previous period. The next column, showing household debt relative to total financial assets, indicates how this expansion of debt has been collateralized—by a rise in asset values rather than incomes. Thus, we see that the liability/asset ratio of American households has not risen at all during the Clinton Presidency, even while the debt/income ratio was shooting up. But the composition of household assets has changed markedly. Traditionally, American property-owners have maintained a steady share of their holdings in insured bank deposits and non-defaultable Treasury securities—prior to the Clinton period, somewhere between 25–27 percent. Under Clinton, this ‘safe asset’ proportion has fallen to 17.8 percent, a stark departure from previous patterns.footnote¹⁶

Finally Table 5 reports figures on real interest rates for 10-year Treasury bonds. It shows that rates did fall in the Clinton period relative to Reagan/Bush years, from an average of 5.5 to 3.7 percent. But the 3.7 percent rate under Clinton is still far higher than the level of any previous presidential era. Indeed, for the whole post-war period 1947–79, the average real Treasury rate was 1.2 percent, less than a third of its level in the Clinton period.footnote¹⁷ These figures make it difficult to argue that the sharp increase in household debt is a response to low interest rates. The reality is that these have been low only relative to the unprecedented peaks of the Reagan/Bush years: they are high by any other historical benchmark. The basic justification given by the Clinton Administration for its drive to eliminate the Federal deficit was that this alone could cut interest rates dramatically, by reducing aggregate demand for credit and enabling the Federal Reserve to pursue a looser monetary policy. In practice, however, rates have fallen relative to the Reagan/Bush years, when Federal deficits soared, but they remain historically high despite the attainment of fiscal balance. The claim that government deficits alone have been responsible for high real interest rates since the 1980s clearly needs to be rethought.

Conditions for workers and the poor

Finally, how have working people and the poor fared during Clinton’s Presidency? Table 6 suggests some measure of their fate. The results are highly unfavourable to Clinton. Despite the relatively strong macro performance—to say nothing of the stock-market boom—both the average wages for non-supervisory workers and the earnings of those in the lowest 10th percent decile of wage distribution not only remain well below those of the Nixon–Ford and Carter Administrations, but are actually lower even than those of the Reagan–Bush years. Moreover, wage inequality—as measured by the ratio of the 90th to 10th percent decile—has increased sharply during Clinton’s tenure in office, even relative to the Republican heyday of the eighties.

Nor has there been any significant reduction in poverty under Clinton, even relative to the Reagan–Bush years. If low rates of unemployment have been a positive feature of the 1990s, it is still quite possible that the overall condition of the poor will prove to have worsened in Clinton’s final years of office, as the dismantling of Federal welfare programmes proceeds. We do not observe any noticeably different patterns if the data are divided by business cycles rather than presidential periods. But we do know that the well-being of working and poor people alike declines during recessions, as unemployment rises and wages fall. This is why the figures for 1991–98, which include a recession, show lower wage rates and higher levels of poverty than during Clinton’s term of office. Had he presided over an interlude of recession, the record would have been even more severe.

Still, whatever else may be said of macroeconomic performance under the Clinton Presidency, the simultaneous fall of unemployment and inflation has defied the expectations of virtually all orthodox economists. In 1999, according to official figures, some 4.2 percent of the work-force were jobless, while inflation was running at 2.4 percent—higher than the 1.6 percent rate for 1998, but otherwise lower than all but two other years since 1965. Most economists, adhering to the Natural Unemployment/Non-Accelerating Inflation Rate of Unemployment doctrines dominant since the early 1970s, had long predicted that unemployment in the region of 4 percent must lead to headlong inflation. They argued that policy-makers therefore had a duty to maintain unemployment at its NAIRU rate—that is, the level below which inflation would take off. To this end, it was generally believed that unemployment needed to be perhaps as high as 6 percent.

What happened to the inflation/unemployment trade-off?

What caused the dramatic shift in the trade-off between unemployment and inflation, and to what extent has the Clinton Administration been responsible for it? Some leading economists have begun to concede that the NAIRU is subject to change over time. Robert Gordon, for one, has concluded from an extensive econometric analysis of the past two decades that NAIRU is ‘time-varying’—falling, for example, from 6.2 percent in 1990 to 5.6 by mid-1996.footnote¹⁸ Douglas Staiger, James Stock and Mark Watson concur, finding that NAIRU in 1997 was between 5.5 and 5.9 percent, a full percentage point below its level for the early 1980s. They also admit that ‘the most striking feature of these estimates is their lack of precision’.footnote¹⁹ Their NAIRU estimate not only varies over time but also has the capacity to range widely at a given point in time.

The general thrust of these broad econometric findings appears solid enough. Indeed, they are difficult to dispute precisely because they are so broad. But in focusing exclusively on point estimates, confidence intervals, and their variation over time, they miss the fundamental question that leaps out from these results—namely, what makes a ‘time-varying’ NAIRU vary in the first place? It is remarkable that leading economists who have devoted so much time to estimating values for NAIRU almost completely neglect this question. Occasionally, however, a few revealing hints are dropped as asides. Gordon, for example writes:

The two especially large changes in the NAIRU . . . are the increase between the early and late 1960s and the decrease in the 1990s. The late 1960s were a time of labor militancy, relatively strong unions, a relatively high minimum wage and a marked increase in labor’s share in national income. The 1990s have been a time of labor peace, relatively weak unions, a relatively low minimum wage and a slight decline in labor’s income share.footnote²⁰

Gordon also casually refers to intensified world competition in product and labour markets, and increased flows of unskilled immigrant labour into the United States, as factors contributing to a declining NAIRU. Though again these observations are mere asides in Gordon’s paper, the overall point is clear: it is changes in the balance of forces between capital and labour, and the growing integration of the US into the global economy—which has made it more difficult for US firms to raise prices and US workers to improve wages—that have been the main forces driving the NAIRU down. Gordon’s general hunch is fully consistent with the econometric results generated by Cara Lown and Robert Rich of the New York Federal Reserve Bank. They found that, between 1990 and 1995, the stagnation of wages and benefits by itself fully explains the lack of inflationary pressure at low levels of unemployment.footnote²¹ Data for the Lown and Rich study end in 1995. Since then, additional factors have contributed to dampening inflation. For one, energy prices fell substantially over 1997–98. In addition, the East Asian financial crisis triggered currency devaluations throughout the region, making American imports cheaper. For their part, workers—especially in the computer industry—have been increasingly willing to follow their employers in taking stock options as part of their total compensation package. That may also be reducing any upward wage pressure, though we should remember that only 21 percent of Americans own any equity outside their share of retirement funds. Finally, of course, we do indeed observe dramatic price increases in this business cycle, if at the stock exchange rather than the supermarket.

The central fact remains, however, that wage gains during the Clinton boom have remained well below those of any other expansion, much less a period of near full employment. This underlying reality is captured in a Business Week report of December 1999 that substantial majorities of US citizens expressed acute dissatisfaction with various features of their economic situation. For example, 51 percent of American workers who were interviewed declared that they ‘felt cheated by their employer’. When asked their view of what Business Week termed the ‘current productivity boom’, 63 percent said that the boom has not raised their earnings, and 62 percent that it had not improved their job security.footnote²² Such negative popular reactions are very striking, given the persistent portrayal by the media of the Clinton economy as a time of unparalleled prosperity. Behind them lies the primary explanation for the collapse of the trade-off between unemployment and inflation, openly acknowledged by Alan Greenspan in his regular semi-annual testimony to Congress in July 1997. Saluting the economy’s performance that year as ‘extraordinary’ and ‘exceptional’, he remarked that a major factor contributing to its outstanding achievement was ‘a heightened sense of job insecurity and, as a consequence, subdued wages.’footnote²³ This ‘heightened sense of job insecurity’ lies at the very foundation of the Clinton administration’s economic legacy.

The stock market boom

The stock market bubble has been the other extraordinary development associated with the Clinton Presidency. What makes it extraordinary is the effect it has had outside Wall Street, on the American and world economy. Dean Baker has summed up its impact as follows:

The run-up in stock prices, in excess of GDP growth, has added more than $8 trillion in financial wealth over the last nine years. A conventional rule of thumb is that $1 of stock wealth increases consumption by 3 cents. This calculation would imply that the $8 trillion of excessive stock market accumulation over the last nine years has increased consumption by $240 billion compared with a situation where the stock market had only kept pace with GDP. This additional consumption corresponds almost exactly to the 4.5 percentage point drop in the saving rate that the economy has experienced during this period.footnote²⁴

The rise in debt-financed consumption has, in turn, maintained a buoyant level of aggregate demand in the US economy, despite the fact that government expenditures have declined and the trade deficit has grown. At the same time, as we have seen, the Federal Government received nearly $50 billion more in revenue in 1997 relative to 1992 from capital gains taxes—by far the largest proportional increase from any fiscal source. Thus the stock market boom has been central both to the creation of a fiscal surplus under Clinton and—through wealth-driven increases in consumption—to counteraction of the negative effects of that surplus on aggregate demand. The boom has, moreover, enabled firms to meet pension fund obligations by rising portfolio values rather than transferring revenues into retirement funds. This in turn has released internal cash-flow for distribution as dividends to shareholders or investment in new capital. Rising share prices have also fuelled the pace of corporate mergers, by enabling firms to buy other companies through stock transfers rather than having either to borrow or pay cash.

Conventional explanations of the bubble give pride of place to the dramatic advances in computer and internet-related technology, which are held to have engendered formidable productivity gains. But we have seen that productivity has not registered exceptional growth under Clinton, even after upward revision of national accounts to make special provision for computer-driven improvements. It is true that there has been a spurt to an annual average rate of 2.6 percent between 1996 and 1999, as against the dismal 0.8 percent of 1993-95. But such productivity figures are hardly a sufficient basis to underwrite the Clinton stock boom—at the much higher annual rate of productivity growth of 3.4 percent in the Kennedy/Johnson period, nothing close to this speculative spree occurred. Moreover, current research casts doubt on the magnitude of the recent spurt itself.footnote²⁵ Of course, the promise of future internet-led leaps in productivity remains. But even if we allow that possibility, it still does not explain the magnitude of the current stock price inflation. As Doug Henwood notes:

The Internet stocks that have headlined the mania over the last year are without known precedent in US financial history. At its highs in early April, the market capitalization of Priceline.com, which sells airline tickets on the web and has microscopic revenues, was twice that of United Airlines and just a hair under American’s. America Online was worth nearly as much as Disney and Time Warner combined, and more than GM and Ford combined. Yahoo was capitalized a third higher than Boeing, and eBay nearly as much as CBS. At its peak, AOL sported a price/earnings ratio of 720, Yahoo! of 1,468 and eBay of 9,571 . . . Oh yes, enthusiasts respond, but these are bets on a grand future. But previous world-transformative events have never been capitalized like this . . . RCA peaked at a P/E of 73 in 1929. Xerox traded at a P/E of 123 in 1961. Apple maxed out at a P/E of 150 in 1980. And all these companies were pretty quick to turn a profit, and once they did, their growth rates were ripping. In the so-called Nifty Fifty era of the early 1970s, the half-hundred glamour stocks that led the market sported P/Es of forty to sixty . . . And those evaluations were once legendary for their extravagance.footnote²⁶

Causes of the bubble

Given the historically unique character of the bubble of the 1990s, it will be some time before we have a definitive account of its causes. But for the moment, and still to some extent groping in the dark, we may point to five significant factors:

1. Financial deregulation. Kindleberger and others have amply documented the way in which speculative manias have historically recurred in financial markets.footnote²⁷ After the Wall Street Crash of 1929 and the slump of the 1930s, post-war governments in all major capitalist economies set in place far-reaching systems of financial regulation to prevent renewed bouts of destructive speculation. In consequence, for the first 25 years after the end of World War II, stock markets were relatively tranquil. This experience suggests one simple explanation for the Clinton boom: that in the absence of effective regulation, speculative excess will inevitably occur in financial markets, though exactly how bubbles will emerge and develop can never be known in advance. In this sense, asset inflation has broken all bounds in the 1990s because the Clinton Administration has operated no adequate controls to inhibit its development.

2. Increased inequality and profitability. As we have seen, the rewards of economic growth under Clinton have been claimed increasingly by the wealthy. Wages have continued to stagnate or decline for most workers, even as GDP and productivity growth have risen. With wages held down as output and productivity rise, profits inevitably increase. Under Clinton they have reached a thirty-year peak. In 1997 the share of total corporate income accruing to profits was 21.6 percent, as opposed to cyclical highs under Nixon (1973) of 18.0 percent, Carter (1979) of 17.4 percent, and even Reagan (1989) of 18.4 percent.footnote²⁸ The escalation of profits under the Clinton Presidency in turn feeds expectations of further increases in profitability, in conditions where the political system continues to favour so heavily the interests of the rich, regardless of whether there are Democratic or Republican incumbents in the White House. If Clinton’s tax policies are less regressive than Reagan’s, they are more so than Nixon’s, while in the areas of trade, financial and even labour markets, the trend of his Administration has been strongly pro-business.

3. Changes in US wealth-holding patterns. We have seen the extent to which American households have moved their portfolios out of low-risk bank deposits and Treasury securities into riskier assets—above all equities. The rise of mutual funds and derivative markets, through which the risks associated with stock-ownership are spread, has certainly contributed to this shift. But it also suggests that property-owners have come to believe that equities are now less of a hazard than they have been at any prior point in history.footnote²⁹ The Clinton Administration alone is obviously not responsible for creating this state of mind among investors. In part, such thinking stems from the rise in profitability and, especially, the positive feedback effects of favourable returns on investor expectations. Alan Greenspan himself has repeatedly tried to dampen such ‘irrational exuberance’ among wealth-holders. But the enthusiasm with which the Federal Reserve and the Clinton Administration have pushed for the deregulation of financial markets has more than counterbalanced any downward jawboning efforts by Greenspan.footnote³⁰

4. Shifts in foreign wealth-holding patterns. From 1989 onwards, the US has become a net debtor nation, as foreign-owned assets in the country have exceeded American-owned assets abroad. Through the 1990s, foreign wealth-holders have increasingly purchased dollar-denominated assets in US financial markets. By the end of 1998, the magnitude of the foreign debt had reached $1.5 trillion, equal to 18 percent of GDP—tripling in size over the previous 24 months.footnote³¹ This inflow of foreign savings is, of course, the other side of the persistent American deficit. Indeed, it is the continued willingness of foreigners to accept payment in dollars and to invest in dollar-denominated assets that alone has made it possible. Here the instability of stock markets across the rest of the world has been critical in making American assets so attractive. Independently of the Wall Street bubble, foreign investors prefer US bonds as well as stocks not because returns on them as such are always highest, but because they are perceived as the best risk-adjusted choice. At the same time, the main source of the rise in foreign-owned assets in the US in 1998 was not an increase in net new holdings, but rather price increases in the value of foreign-held American assets relative to the prices of American-held foreign assets.

   Overseas savings in US financial markets have in turn increased total demand for US securities and thus, all else being equal, their price. The effect here has primarily been indirect, since the share of foreign ownership in total American stock market capitalization has held fairly constant throughout the 1990s, at around 8 percent. But, as Jane D’Arista argues, because increased foreign purchases of US bonds have pushed their prices up and yields down, they have encouraged domestic investors to switch into stocks. ‘US equity markets could not have risen so far so fast without the benefit of substitution effects from large capital inflows.’footnote³²

5. Adept Federal Reserve policy. The Federal Reserve has been praised for allowing unemployment to fall well below the level that NAIRU hawks had said was prudent. But, as we have seen, Greenspan understood that job insecurity would inhibit American workers from pressing for wage demands even in tight labour markets, as they had done in the past. His real achievement during the Clinton presidency has lain elsewhere—in holding a balance between the need to keep financial markets liquid enough to sustain the stock market, and to keep interest rates high enough to ensure a continued flow of foreign savings into the US. Greenspan has certainly managed this well, even as the countervailing market pressures have mounted. Furthermore, had Greenspan and Rubin not conducted successful bail-out operations when the sequence of Mexican, East Asian and Long-Term Capital Management crises broke out, the US stock market would probably have dived as the cumulative effects of these shocks coursed through global financial markets.footnote³³ By a ‘successful’ bail-out, I mean an operation that not only prevented an interactive debt deflation, but also protected the wealth of US investors—since substantial losses would almost certainly have burst the US bubble.