The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
For release on delivery 12 30 p m E D T June 6, 1996 Remarks byAlan Greenspan Board of Governors of the Federal Reserve System at the 40th Economic Conference of the Federal Reserve Bank of Boston Chatham, Massachusetts June 6, 1996 I regret that I was unable to join you for the earlier portions of this conference I know that you've had some important explorations of the process through which technology contributes to economic growth What I'd like to do in this session is perhaps augment these discussions by shifting gears a bit I would like to focus on the question of how people perceive the benefits of recent technological change Today a truly puzzling phenomenon confronts the American economy I refer to the pervasiveness of job insecurity in the context of an economic recovery that has been running for more than five years, inflation that has been contained, and a layoff rate that is historically quite low Yet, in the face of all this seemingly good news, a sense persists that something is fundamentally wrong This afternoon I want to try to explain where I believe the insecurity is coming from and, I hope, raise some suggestions as to how it might be assuaged The issue, as best I can judge, appears to be rooted in one of those rare, perhaps once-in-a-century events--a structural technological advance The advent of the transistor and the integrated circuit and, as a consequence, the emergence of modern computer, telecommunication, and satellite technologies have fundamentally changed the structure of the American economy Since the beginning of the industrial revolution, our economy and, to only a slightly lesser degree, the economies of - 2 our industrial trading partners have been progressing toward a regime in which abstract ideas and concepts are the dominant element in the creation of economic value A hundred years ago, physical brawn was critical to value-added determination People who personally could lift rolled sheet steel and help haul it from one part of the plant to another performed an activity that was valuable in the marketplace Today, several generations later, the structure of production has become, to a remarkable degree, idea-determined On the output side, at the turn of the twentieth century, we produced steel, industrial chemicals, and heavy fabrics in abundance, what impressed was the very size and bulk of the productive facilities and the output itself Today, the products that we find remarkable are those that are lighter, smaller, and in some cases, almost invisible Our radios used to be activated by large vacuum tubes, today we have pocket-sized transistors to perform the same function huge tonnages of copper wire Thin fiber optic cables have replaced In the past, buildings were so over-structured and sturdy that, when their time for replacement arrived, demolition was a Herculean task Owing to conceptual advances in metallurgy, engineering, and architectural design, we now can create as much or more space with fewer materials Indeed, such advances have created an overall national output whose physical weight probably is only modestly greater than that of whatever we produced a hundred years ago Real GDP, - 3 that is price-adjusted value-added, of course, is much higher today, and by far, ideas account for the difference That trend will doubtless continue because idea creation is irreversible Knowledge, once acquired, does not disappear If anything, this process has accelerated in recent years, and that acceleration seems to have had two important side effects First, it has had a major influence on the distribution of income in this country, and second, a related but different concept, it has imparted a degree of insecurity, uncertainty, and even fear to a vast segment of job holders The consequence of both effects, as I will explain shortly, has been to create a sense that something in the economy is awry, which is wholly at odds with what the macroeconomic data seemingly imply--economic success, tranquility, and progress The roots of this puzzling situation go back a few decades As ideas became especially valuable relative to physical activity in the creation of value-added, education and intellectual skill became increasingly major determinants of income Throughout the 1960s and 1970s, the rapid rise in the number of college graduates apparently kept the supply of educated workers moving up with the demand However, by the latter 1970s and into the 1980s, demand seemed to have outstripped supply, the apparent consequence was a fairly pronounced rise in compensation going to college graduates relative to the compensation going to those who had only high school diplomas A similar disparity of earnings - 4 developed between those who had graduated from high school and those who had dropped out After the mid -1970s, productivity slowed quite markedly, for reasons that are not wholly apparent, and so did average real incomes As a consequence, the widening disparity also means that a not insignificant portion of our work force--primarily those whose work involves less conceptual activities--has been experiencing either stagnant or falling real incomes in the past ten or fifteen years A substantial number of these people understandably feel that they have been on a treadmill and are barely able to make ends meet from their incomes That feeling has engendered significant concerns about economic and financial well-being among this part of our work force I suspect that other concerns affect an even larger group--composed of those who have average or above incomes and have been employed in their current jobs for a number of years These are the people with higher skills, who interact closely day by day with the high-tech part of our capital stock Because that stock, reflecting computer and telecommunications-based technologies, is turning over very rapidly, the involved workers have a high degree of uncertainty and insecurity about their jobs As one affected employee commented to a Wall Street Journal reporter a couple of weeks ago, "Is somebody getting - 5ready to change my whole life for me?" 1 These workers perceive the job skills that they have acquired through high school or college to be increasingly open to competitive challenge One must wonder how highly skilled, turn-of-the-century telegraphers felt with the onset of the telephone or the skilled buggy-whip craftsman with the advent of the automobile Today, large numbers of people have become so demonstrably insecure about whether their skills will still be relevant in, say, five years that they fear for their jobs This insecurity is evidenced by the fact that they have increasingly forgone wage hikes for job security As a consequence, the past few years have been a period of extraordinary labor peace In fact, 1995 had the lowest strike record for a half-century Moreover, labor contracts, which historically almost never extended beyond thirty-six months, are now sometimes going out five and six years, as people try to lock in job security, often willing to forgo significant wage increases in the process This sense of job insecurity is so deep that many workers are truly scared Some fear that their skills will no longer be appropriate for the future meet in the future Some fear their ability to make ends Many appear truly concerned about a prospective decline in their standard of living Wall Street Journal, May 16, 1996, page A 16 - 6 This development is startling considering the overall state of the economy suggested by the macroeconomic data It is certainly the case that average real income has slowed and that the disparity in real incomes has widened After reaching a postwar low in the late 1960s, income disparities, as measured by Gini coefficients, climbed steadily through 1994--the most recent year for which data are available Moreover, disparities in the distribution of wealth (net worth) as measured by the Federal Reserve's Survey of Consumer Finances also widened significantly between the surveys taken in 1963 and 1992, with much of that increase in Gini coefficients occurring during the 1980s Doubtless, that disparity has widened further in recent years in the wake of major increases in stock and bond prices But the notion that the economic well-being of the lower-income segments of our workforce has deteriorated as much as might be suggested by the widening disparities in the income and wealth statistics is open to question I say this because there is a surprising difference between trends in the dispersion of holdings of claims to goods and services (that is, income and wealth) and trends in the dispersion of actual consumption, which is, of course, the ultimate determinant of material or economic well-being Put another way, well-being is determined by things people consume, either directly from their incomes and accumulated savings or indirectly from the stock of household goods they already own--automobiles, telephones, TVs, VCRs, and so forth, not to - 7 mention the homes themselves And disparities in consumption and ownership of hard goods don't appear to have widened nearly as much as income disparities I do not wish to disparage income as a partial antidote to insecurity Nevertheless, some aspects of economic well-being may be more accurately discerned by examining consumption A number of researchers have compared trends in the distribution of consumption with the distribution of income Many of these studies rely on data from the Consumer Expenditure Survey that the Bureau of Labor Statistics conducts, and much of the analytical research on distributional issues has been carried out by BLS economists A recent study by David Johnson and Stephanie Shipp of the BLS finds that "income inequality is more volatile than consumption and the level is about 3 0 percent more than that of consumption inequality "2 These findings are not surprising As is well known, consumers tend to maintain their levels of consumption in the face of temporary changes in income Variations in asset holdings and debt buffer changes in income In short, consumption patterns tend to look more like patterns in income that has been averaged over several years, rather than the oneyear convention of our statistics 2 David Johnson and Stephanie Shipp, "Changing Inequality in the U S from 1980-1994 A Consumption Viewpoint," manuscript, Bureau of Labor Statistics, January 1996, and U S Department of Labor, Report on the American Workforce, 1995 But, besides finding differences in the levels of consumption and income inequality, Johnson and Shipp find differences in the inequality trends In particular, although consumption inequality has increased, on average, since 1981, the rise has been only three-fourths as large as that of income inequality (see table 1) 3 An evaluation that views consumption not in terms of outlays but, rather, in terms of the flow of services that come from purchases indicates an additional qualification The reason, of course, for examining the flow of services from spending, and not just current-period spending alone, is that while outlays for food and haircuts, for example, are consumed immediately, a television set that is purchased today provides entertainment over its entire service life Thus, unless ownership of household appliances and other consumer durables is brought into the evaluation, the story of the dispersion of material wellbeing is incomplete What do the numbers show? During the 1960s and 1970s, the real net stock of consumer durables per household increased an average of 3 1 percent per year The average growth rate has slowed slightly since then--to a pace of 2-1/2 percent--but all of that slowing occurred during the recessions of 1980 and 1981-82 Indeed, since 1982 households have been adding to their 3 The Gini coefficients for consumption and income that are reported on table 1 are based on annual average data - 9 stock of durables at an annual rate per household of 3 3 percent--slightly faster than in the 'sixties and 'seventies 4 Moreover, we have apparently not had a widening disparity in holdings of hard assets like the one that appears in the income and wealth data Stephanie Shipp and her colleagues in the Division of Consumer Expenditure Surveys at the BLS generously provided the Board's staff with detailed tabulations of the ownership of consumer goods and vehicles by income decile To be sure, these data show that ownership rates for consumer durables clearly rise with income But the data also show that for motor vehicles and a number of appliances--for example, dishwashers, clothes dryers, microwave ovens, and even garbage disposals--the distribution of ownership rates by income decile moved toward greater equality between 1980 and 1994 (see table 2 ) 5 For some consumer goods we are moving toward greater equality because the proportion of households with access to these items is moving close to saturation For example, nearly all poor families have access to a refrigerator, stove, and color 4 The growth rate of the net stock of owner-occupied housing (measured in 1992 dollars) per household was 2 3 percent annually from 1959 to 1979, 1 3 percent from 1979 to 1994, and 1 8 percent from 1982 to 1994 5 The calculation of the measure of distributional inequality used to support this statement is described in the attached technical note - 10 TV In addition, three-fourths of poor households have telephones, and nearly two-thirds have microwave ovens and VCRs 6 These encouraging findings are not without qualification, however As an example, for personal computers, which nowadays are critical for economic success, the disparity in ownership rates is quite large--around 10 percent for lower-income households in 1994 compared with more than 50 percent for the highest-income decile And, even when most families own a durable good or vehicle, the number owned by the low-income group typically is less than that owned by the upper-income groups For example, in 1994 lower-income families owned slightly more than one color television set, on average, whereas high-income families tended to own more than two The figures for motor vehicles are similar--slightly under one per household at the lower end of the income distribution and slightly more than two at the upper end Nonetheless, even though the inequality in the number of units owned per household is often greater than that in the ownership rate, the degree of inequality measured on this basis narrowed between 1984 and 1994 in a manner similar to the shifts for ownership rates (compare tables 2 and 3 ) 7 6 Some of these data are taken from Kathleen Short and Martina Shea, "Beyond Poverty, Extended Measures of Weil-Being 1992," U S Bureau of the Census, Current Population Reports, P70-50RV, November 1995 7 Collection of data in the Consumer Expenditure Survey on the average number of units owned per household did not begin until 1984 - 11 But, even if the number of hard assets per family were the same for rich and poor, it is not evident how much this would assuage the current deep-seated sense of insecurity that pervades such a large segment of our workforce Clearly, there is more to economic security than owning consumer durables In fact, the very forces that load our households with every sort of gadget come from an economy that apparently is changing too quickly for many Americans to absorb readily fear in all walks of life Accelerated change fosters It is a rational human response to such an imperative Finding a solution to such insecurity is not simple If 30b insecurity is largely a fear of skill obsolescence, real or imagined, some way must be found to enhance skills People who believe that their skills are up to date and readily marketable do not inordinately fear job layoffs Bolstered by signals from the marketplace, education is clearly increasingly becoming a lifetime activity Resting on one's skills as the world rapidly goes by will only intensify a sense of job insecurity Ongoing schooling and training are becoming ever more relevant for the average worker Fortunately, developing human capital is rapidly being perceived by many corporations as adding to shareholder value If ideas are increasingly the factor that engenders value-added, then training and education are crucial to the expansion of company value-added and profitability - 12 As a consequence, corporate universities are emerging as a growth industry in this country A significant and expanding number of companies require that employees attend class, say, twice a week, at company expense, to augment their on-the-job techniques Moreover, there is a growing peripheral industry whose basic product is to train company employees in the latest technologies Such trends should decidedly be encouraged Hopefully, in that environment efforts to increase the competitive skills of workers in the lower half of the income distribution will succeed in narrowing income disparities * * * At this point it is unclear whether the particular current surge of technology is peaking and will eventually slow down or whether we are in its early stages Much of this surge may well represent more wheel-spinning than real increases in production, as our subdued national productivity data suggest Nathan Rosenberg in his paper for this conference points out that organizational changes and further development of complementary technologies likely will be required before we see the productivity payoff to computer technology If so, as the infrastructure of the economy finally adjusts itself to the new semiconductor-based revolution, the rapid changes are likely to finally become more evident in increased measured productivity and growth - 13 In any event, a new world is emerging The twenty-first century will be different--much more rapidly paced and changing than any of us who have been around for a while have experienced in our lifetimes There will be a different America out there Fortunately, job insecurity does not appear to be a problem for a 21-year-old who has experienced nothing else, and even less for a 6-year-old who seems to be far more computer literate than grandfather As a consequence, with the inexorable turnover of the population, people will adjust When we go through a period of transition, inevitable symptoms of friction, uncertainty, and fear arise They will pass - 14 Technical Note The raw data on the ownership rates of consumer durables by income decile are not in a form that can be used directly to calculate standard measures of inequality ( e g , Gini coefficients or mean log deviations) However, William Cleveland of the Board's staff suggested a transformation of the raw data that allows one to calculate a measure of inequality that looks like a Gini coefficient This note describes the procedure The first step is to transform the raw data into a discrete probability distribution In the case of ownership rates for consumer durables, the calculation for a given consumer good is where p1 is the fraction of all households that own the consumer good who are in income decile 1, and r1 is the actual ownership rate for the 1th decile By construction, the sum of the p1' s is equal to one For goods that have ownership rates that are relatively equal across deciles (regardless of the level of the ownership rate), these probability distributions are fairly flat, with values of p1 close to 0 1 For goods that are more concentrated among the affluent households, the probability distributions tend to rise across income deciles The next step is to take the probability distributions and create cumulative probability distributions (CPD) (e g , the value of the CPD for the second decile equals p1 + p2) The CPDs look like Lorenz curves The standard formula for the Gini coefficient is then used to construct a measure of the degree of inequality implied by the CPDs 8 These are shown in table 2 The calculation of "Gini coefficients" for the average number of units owned per household in each income decile (u1) is the same, except u1 is substitued for r1 in equation 1 These "Gini coefficients" are shown in table 3 8 The "Gini coefficient" is defined as one minus twice the area under the CPD Although this statistic looks like a Gini coefficient, it doesn't have all the properties of a true Gini coefficient For example, a true Gini coefficient must fall between zero and one, but the "Gini coefficient" calculated here could have turned out negative if, say, poor people had owned more microwave ovens than rich people - 15 - TABLE 1 GINI COEFFICIENTS FOR CONSUMPTION AND INCOME Year Consumption Income 1980 0 291 0 365 1981 0 286 0 369 1982 0 299 0 380 1983 0 298 0 382 1984 0 307 0 383 1985 0 315 0 389 1986 0 326 0 392 1987 0 322 0 393 1988 0 320 0 395 1989 0 325 0 401 1990 0 325 0 396 1991 0 321 0 397 1992 0 331 0 403 1993 0 321 0 429 1994 0317 0 426 Sources Consumption data are from the Consumer Expenditure Survey, income data are from the Census Bureau - 16 - TABLE 2 GINI COEFFICIENTS" FOR OWNERSHIP RATES OF SELECTED CONSUMER DURABLES (By income decile) Microwave ovens Dishwashers Clothes dryers Garbage disposals Motor vehicles Freezers Clothes washers Refrigerators Stoves 1980 1994 28 29 17 26 09 06 08 01 01 08 22 12 19 07 07 09 01 01 Source Based on tabulations from the Consumer Expenditure Survey See the technical note for a discussion of the method used to calculate the Gini coefficients TABLE 3 "GINI COEFFICIENTS" FOR NUMBER OF UNITS OWNED PER HOUSEHOLD OF SELECTED CONSUMER DURABLES (By income decile) Microwave ovens Dishwashers Clothes dryers Garbage disposals Motor vehicles Freezers Clothes washers Refrigerators Stoves 1984 24 27 15 23 14 06 08 03 1994 08 21 12 19 13 07 09 02 02 Source Based on tabulations from the Consumer Expenditure Survey See the technical note for a discussion of the method used to calculate the Gini coefficients
@FedFRASER