THE REAL ROOTS OF THE GREAT RECESSION UN

DIPARTIMENTO DI SCIENZE ECONOMICHE AZIENDALI E STATISTICHETHE REAL ROOTS OF THE GREAT RECESSION: UNSUSTAINABLE INCOME DISTRIBUTION MASSIMO FLORIO Working Paper n... MASSIMO FLORIO* THE

DIPARTIMENTO DI SCIENZE ECONOMICHE AZIENDALI E STATISTICHE

THE REAL ROOTS OF THE GREAT RECESSION:

UNSUSTAINABLE INCOME DISTRIBUTION

MASSIMO FLORIO

Working Paper n 2012-01 FEBBRAIO 2012

MASSIMO FLORIO*

THE REAL ROOTS OF THE GREAT RECESSION: UNSUSTAINABLE INCOME

DISTRIBUTION

Abstract: This article suggests that, at the root of the Great Recession, there are two imbalances: an

unsustainable distribution of income in advanced economies, and a growing income gap between them and the rest of the world Several factors have squeezed the share of labour income, particularly of less skilled workers in advanced capitalist economies and hugely increased the share of capital income (sometimes disguised as pay for the work of top managers) In the US and elsewhere, the demand for consumer goods did not fall until the recent Great Recession thanks to the growth of private debt A progressive income redistribution policy is seen as the core structural reform needed

to rebalance the system, along with a change in international economic relations

Keywords: Great Recession, income distribution, shares of labour and capital income

* A revised version will appear in International Journal of Political Economy

1 Introduction: Regulatory or Market Failures?

This paper suggests that, while in the short run only a combination of monetary and fiscal policy can counteract the current crisis1, the long-run cure needs to address the core structural imbalances: an unsustainable distribution of income in advanced economies, and a widening income disparity between them and the rest of the world Various interpretations of the Great Recession circulate among economists and politicians that spawn a number of short and medium-term cures2 I shall label them the ‘conservative’, the

‘progressive’, and the ‘alternative’ views

The first thesis, favoured by market fundamentalists and other conservative economists, is that the underlying reasons for the crisis are an excess of public interventionism, especially in housing policy, the regulation of financial markets, and loose monetary policy The conservative economists claim that for demagogical reasons the US policy makers forced the financial system to grant home mortgages to poor people who could not afford them, sparking off a cycle of increasing house prices that was totally artificial and could only burst eventually In addition, although aware of the ‘bubble’, the Federal Reserve oscillated

in its monetary policy, first by substantially reducing the cost of money for the banks, and then abruptly raising it again, thus providing both the oil to start the fire and the cold shower to put it out, but causing a flash flood instead In short, always for reasons of political opportunism, different US governments and the Congress, officially or unofficially, led the financial system to believe that the excessive risks it was taking would have been covered by various forms of public guarantees The corollary of this view is that governments should now avoid intervening and they should let the crisis take its course in order to find a new, healthier equilibrium, after having swept away the badly managed companies

My preferred example of this interpretation is the opinion by Mulligan (2008), a macroeconomist at

the University of Chicago in an invited editorial in the New York Times, in the early days of the recession,

three weeks just after the crash of Lehman According to Mulligan, “Since World War II, the marginal product of capital, after taxes, has averaged 7% to 8% per year… And what happened during 2007 and the first half of 2008, when the financial markets were already spooked by oil price spikes and housing pricing crashes? The marginal product was more than 10% per year, far above the historical average The third quarter earnings reports from some companies already suggest that America’s non-financial companies are still making plenty of money… So if you are not employed by the financial industry (94% of you are not), don’t worry The current unemployment rate of 6.1% is not alarming…” (Mulligan 2008: A33) The core

tenet of this way to react to the then incoming storm is the optimistic interpretation of the “far above the

historical average” marginal product of capital in the non-financial sector

A sample of additional themes and arguments in this market fundamentalist view of the crisis are given by Smith (2008), according to whom there is nothing new in financial bubbles, which share a simple feature: they come to an end by themselves, and nobody knows how to cure them, until they simply burst In Smith’s view the origin of the crisis is related to a giant real estate bubble, more than a security bubble In turn, the expectations of price increase in the real estate market has been fuelled by legislation in favour of home ownership, particularly the Tax Relief Act of 1997, that exempted from taxation, up to 500,000 USD the capital gains from reselling a home after at least two years

In the same vein, Liebowitz (2007) observes that the real estate market had started to invert its pace already in the second quarter of 2006, and, in mid-2007, it was clear that there was a contagion to the financial sector, and subsequently to the securities market He thus finds evidence of the forecast done much earlier by Dand and Liebowitz (1998), that sub-prime lending, particularly to ethnic minorities, i.e non-discriminatory lending, would have created a solvency problem Liebowitz (2008) shows by simple diagrams

two facts: first, expressed in constant USD (1983), the yearly price of homes between 1987 and 1996 was

slightly decreasing, while it increased in real terms by more than 80% between 1995 and 2005, and then

sharply declined, from 2006 on; second, between 1970 and 1995 the yearly share of households owning a

home had been nearly constant, between 64-65%, while it had peaked to 69% in 2004, and then it had started

to decline again

According to this kind of interpretations, the abnormal trend in the construction industry and in the real estate market would not have been possible without the accommodating role of the Federal Reserve O’Driscoll (2008) is an example of the view that the “Greenspan put” and then the “Bernanke put” in monetary policy are guilty of disturbing the role of prices as the only signal that would help markets to adjust themselves The origins of the sin, according to this view, goes back the federal guarantee on cash deposits in the 1930s, the related moral hazard, and the tendency of non-neutral monetary policy to disturb markets, as pointed out at that time by Friedrich von Hayek and Ludwig von Mises, and later on by Milton Friedman According to O’Driscoll, while the Fed has had over decades an implicit inflation target at around 2% per year, in fact it manipulated the CPI, by for example excluding the prices of food and energy3 In the Hayek-von Mises perspective, any attempt to stabilize the prices of goods and services may help to create an asset bubble See also White (2008) for the view that the recession in the US is intrinsically due to a combined failure of monetary and housing policies

The ‘progressive’ thesis, often labelled as “Keynesian”, argued especially by Paul Krugman and Joseph Stiglitz, is that the financial markets have a historical responsibility in the crisis, basically due to an excessive greed for profit4, without adequate controls, which also brought about great imbalances in the distribution of income The cure should entail expansive fiscal policies, far more courageous than the current ones, because global demand seems insufficient, putting on one side for the moment the problem of the sustainability of the public debt and the risks of inflation

According to Krugman (2009), the Asian crisis of the 1990s, the Japanese asset bubble, or the crises

in Mexico and Argentina had to be considered as a rehearsal for the current troubles, which in turn are related to other histories of the instability of capitalism This is so despite the premature prophecy by Robert Lucas, in his presidential address at the American Economic Association that economists have understood how to tame the business cycle Krugman suggests that the repetition in the US of the episodes of instability,

as already experienced elsewhere, has to be in part attributed to the Fed’s unwillingness under Alan Greenspan to curb ‘irrational exuberance’ in the financial markets As there was a conventional view in the 1990s that inflation would not have been accelerating until the unemployment rate fell below 5.5%, in fact with unemployment going down to less than 4% around the turn of the century, inflation remained quiescent This was attributed by Greenspan and others to fundamental changes in technology and productivity, and no need was seen to raise interest rates According to Krugman “Greenspan warned about irrational exuberance, but he didn’t do anything about it And in fact, the Fed chairman holds what I believe is a unique record among central bankers: he presided over not one but two enormous asset bubbles, first in stocks, then in housing” Under this view, the ‘dot.com’ bubble, and the related euphoric rise of the price of stocks, and the rise of prices of houses were a combination of irrational expectations, wrong monetary policy, and greed of

therapy According to Krugman, the origin of the crisis is to be traced back to the irrational functioning of

the unregulated financial markets To simplify, the conservative view points to a regulatory failure in the first place that gave rise to a bubble Krugman’s view points to a market failure, not addressed by regulation

Hence the policy prescriptions of the two views are diametrically opposed To the conservative economists what is needed is mainly laissez-faire; while, to the progressive economists, what is required is public spending and accommodating monetary policy in the short term and tougher regulation of financial markets

as a structural therapy

Market failures are also central in Stiglitz’s (2010) interpretation Most of his book is devoted to documenting the manipulation of information by different financial players, and how their customers were not protected, as in fact the Fed and policy-making were captured by the interests of the financial industry Stiglitz also suggests that most economists share some responsibility for what has happened, as they have misunderstood the conditions under which the Arrow-Debreu representation of general equilibrium is valid, particularly perfect information and exogenous technical progress Should the economists had read more carefully the Greenwad and Stiglitz (1986) theorems (on markets affected by externalities and incomplete information), they would have realized that only government can cure some fundamental market failures According to this view, the proposed cures are a combination of public spending, regulation of the financial markets, more progressive income taxation, and a dose of industrial policy to sustain technical progress The

“alternative” view departs from both the “conservative” and the “progressive” views, as it suggests a major role for an income distribution shock as an ultimate causal factor, and suggests adding a structural income policy to the short-run fiscal and monetary policy measures

The structure of this paper is as follows: section 2 focuses on income distribution disequilibria as the real roots of the crisis; section 3 focuses on factor shares; section 4 further elaborates on this interpretation; and section 5 discusses possible approaches to decrease the weight of capital income and concludes

2 The Role of Income Distribution

Stiglitz (2010) observes that the median income in the US in 2008 was 4% lower in real terms than in 2000, while the per capita GDP was 10% higher The average is strongly influenced by asymmetric distribution and by the vertiginous growth of high incomes

I briefly elaborate below on this issue, which in my opinion leads to an alternative view of the Great Recession, based on the role of income distribution shocks in the last two decades (see particularly Foster and Magdoff (2009))

35,000 37,000 39,000 41,000 43,000 45,000 47,000 49,000 51,000 53,000 55,000

1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009

1967: USD 40,770

1999: USD 53,252

2010: USD 49,445

Figure 1 Real Median Household Income in the U.S., 1967-2010 (in 2010 USD)

Source: Author’s calculation based on data of U.S Census Bureau (2010), Current Population Survey, Historical

income Tables, Table H-5 Retrieved from

http://www.census.gov/hhes/www/income/data/historical/household/index.html

Legend: Dotted lines show the trends over the period considered, with a structural break around 1999

Figure 1 shows the real median household income in the US, between 1967 and 2010 (the last few years being the most affected by the crisis, with, for instance, a decrease of 4.2% between the years 2007-2009) The figure suggests that 1999 was the peak year, with USD 52,388 income for the median household,

a level that was never to be reached again, and that, in fact, began to decline well before the recent crisis Over this 30-year time span, real family income of the bottom 20% decreased in real terms by 7.4%, and the second 20%, those with income between around 27,000 – 48,000 USD, almost stagnated, with an average increase per year of below 0.1% per year In fact, it seems that for one half of the US households the income

in the last thirty years did not increase on average, while the increase was mostly concentrated in the top 40%, and particularly with the top 5% The contrast with the previous dynamics, as showed in Figure 2, particularly looking at the bottom of the distribution cannot be exaggerated In terms of income distribution Piketty and Saez (2003)5 find that the share of the top 1% of total pre-tax income was at its top, 23.9%, in year 1928, which as it happens was exactly the year before the Great Depression, then steadily declined, was just 8.6% in 1976, and then increased to nearly 1928 levels, 23.5%, in year 2007, again one year before the current Great Recession In both cases, the turning point was a crash of the stock market, but the increase in income inequality was built over a number of years before the crash, see Figure 3

More in general, according to Piketty and Saez (2003) and updates, when adjusted for inflation, between 1973 and 2005, the average income of the first nine deciles fell by 11%, while average productivity increased by 80% According to Baker (2007) the divorce between productivity changes and pay in the last thirty years is in sharp contrast with the productivity and wages in the previous decades, which increased at the same regular pace of 2-3% every year

-20%

0%

20%

40%

60%

80%

100%

120%

Bottom 20%

Second 20%

Middle 20%

Fourth 20%

Top 20% Top 5%

1947-1979 1979-2009

Figure 2 Change in Real Family Income by Quintile and Top 5% in the U.S., 1947-1979 and 1979-2009

Source: Author’s calculation based on U.S Census Bureau data in Lawrence and Bernstein (1994) and U.S Census

Bureau (2010), Historical Income Tables: Families, Table F-3 Retrieved from http://inequality.org/income-inequality/

and http://www.census.gov/hhes/www/income/data/historical/families/index.html

5%

10%

15%

20%

25%

30%

1913 1917 1921 1925 1929 1933 1937 1941 1945 1949 1953 1957 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 2005

1928: 23.9%

1976: 8.9%

2007: 23.5%

2008:

21.0%

Figure 3 Top 1% Share of Total Pre-Tax Income in the U.S., 1913-2008

Source: Author’s calculation based on Piketty and Saez (2003), updated to 2008 Retrieved from on

http://inequality.org/income-inequality/

It seems as the post-war social history of the country can be divided between three decades of cohesive growth, and another three decades of explosive inequality, similar to the situation during the 1920s These trends in income are reflected in their cumulative effects on wealth inequality According to the data reported by IPS from various sources, against a median net worth of US households at 62,000 USD, the top 1% holds 35.6% of total wealth, and the bottom 80% just 12.8% in 2009 In terms of distribution of the US stock market wealth in 2007, while the top 1% holds 38.2%, the bottom 80% owns just 8.9% of the pre-crisis value

Turning to an international perspective, the OECD (2011) observes that growing inequality is a common trend across the advanced economies between the mid-1980s and late 2000s They consider disposable household income, deflated for the consumer price index, and find that in the last 20-25 years the average annual change of income of the bottom decile has been 1.4%, and 2.0% for the top decile Countries that show a much larger gap include United States (0.5 bottom, 1.9 top), Austria, Israel, Italy, Japan, Netherlands and Sweden Typically the ratio of the top to the bottom income is nine times in the OECD, but much lower in the Nordic and other European countries, and much higher in the US (14 times), Mexico and Chile (27 times) Taking the Gini index, it increased in 17 out of 22 countries The OECD report acknowledges that the common trend towards increased inequality started in the UK and in the US in the late 1970s and early 1980s, but it has then been spreading everywhere

Nevertheless, it is difficult to argue that in past years global demand has been weak Consumption in the USA and in the rest of the world (especially in emerging economies) has risen sharply What seems to be the key point is that while in some countries, for example China, India and Brazil, consumption increased

thanks to the sustained growth in household income and saving, in the United States and elsewhere, it increased thanks to the growth in household debt In other words, the problem was not that effective demand

was insufficient, but that it was unsustainable, given the income distribution in some key countries of today’s global economy

3 On Factor Shares

Perhaps less prominent in the debate on the Great Recession has been the subject of the distribution of income among the factors of production Historically, in the last century, except for major upheavals related

to wars and specific economic crises, a sort of social pact had led to an approximately constant ratio in the distribution of income between work and capital out of the gross national product of each country (even though there were oscillations over time and differences between countries) This was one of the stylized facts in Kaldor’s (1961) view of growth, and while the shares can fluctuate year by year, they were assumed

to reverse to a constant mean value This ‘pact’ was broken in the 1980s and the share of income from capital rose abruptly, automatically reducing the share of labor Measurement is difficult in this area of the national accounts (see, for example, OECD (2011)), but the intuition of the overall trend is clear enough

Perhaps the most revealing aspect of this story, is given by Figure 4, which shows average hourly wages, as a crude way to understand the dynamics of labor versus capital income In terms of constant 2009 USD, the wage between 1947 and 1972 increased by more than 75%, while in the following period until

2009 the increase was only 4% (over 37 years), and in fact it decreased in real terms until the late 1990s

At the same time, the federal minimum wage has been declining

2.50 7.50 12.50 17.50 22.50

Average hourly wage

Federal minimum wage

1945:

USD 10.18

1972: USD 17.88

2009: USD 18.63

Figure 4 U.S Federal Minimum Wage (1956 - 2006) and Average Hourly Wage (1947-2009) – US dollars

Source: Author’s calculation based on Boushey and Tilly (2009) and Economic Policy Institute data (2011) Retrieved

from http://inequality.org/income-inequality/

Note: The federal minimum wage is expressed in 2006 constant dollars and the average hourly wage is expressed in

2009 dollars

46%

47%

48%

49%

50%

51%

52%

53%

54%

1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007

Figure 5 Wage and Salary Disbursement as a Percentage of GDP, 1959-2007 (%)

Source: Author’s calculation based on Foster and Magdoff (2009), Chart 3 from Economic Report of the President,

2008 Retrieved from http://monthlyreview.org/2008/12/01/financial-implosion-and-stagnation

Additional evidence is provided by Foster and Magdoff (2009) According to these authors, the divorce between wages and productivity changes is a long term process, with falling wages and salaries disbursements as a share of GDP in the US, between 1960 and 2007 (see Figure 5) In Foster and Magdoff’s words: “… real wages of private non-agricultural workers in the United States (in 1982) dollars peaked in

1972 at USD 8.99 per hour, and by 2006 had fallen to 8.24 (equivalent to the real hourly wage rate in 1967), despite the enormous growth in productivity and profits over the past few decades… The truly remarkable fact under these circumstances was that household consumption continued to rise from a little over 60% of GDP in the early 1960s to around 70% in 2007 This was only possible because of more two-earner households…, people working longer hours and filling multiple jobs, and a constant ratcheting up of consumer debt Household debt was spurred, particularly in the later stages of the housing bubble, by a dramatic rise in housing prices, allowing consumers to borrow more against their increased equity…household debt increased from about 40% of GDP in 1960 to 100% of GDP in 2007, with an especially sharp increase starting in the late 1990s”

The tendency of labor share in national income to fall is not observed in the US alone The trend is common to most major advanced economies This fact is acknowledged by the IMF (2007), in a report especially focusing on the issue of the structural change of the historical pattern of labor/capital shares and increasing income inequality One explanation of this trend points to explicit labor policies in the late 1970s

to stop the increase of wages and to constrain trade unions Additional possible explanations are technological changes, which displace unskilled workers, and the wage competition between workers in developed economies and those in emerging countries, as China, in the context of trade globalization (see OECD (2011)), which also mentions regressive tax-benefit reforms in the last two decades

According to the OECD (2011) most of the household income inequality has been driven by rising inequality in wages and salaries, as they typically account for 75% of incomes of working adults Atkinson (2009) observes that top earners have experienced a particularly fast growth What seems difficult, however,

is to disentangle capital and labor income of top earners Moreover, the role of self-employment poses additional problems of interpretation It is disproportionately represented in the bottom household income quintile, but also in some business in the top quintile, such as doctors and lawyers, where capital income is often combined with labor income Focussing on capital income, the OECD (2011) suggests that the share of capital income remains modest in average household income, typically 7%, and “not surprisingly where this

share increased, it was predominantly due to movements in upper part of the distribution”

According to Thompson and Smeeding (2011), the evidence points to the following:

a) Until 2007 and starting from 1979, the unemployment rate of less than high-school achievers shows oscillations between a maximum of around 16% in 1983, and a minimum of around 8% in 2006 A similar cyclical pattern is shown by those with high school only, but with a lower average unemployment rate (10 to 4% range), while except in post recession years, those with a bachelor or/and advanced degree have typically experienced a much flatter unemployment rate, between 2 and 4% This points to a high responsiveness of unemployment of ‘blue collars’ to output shocks, which is well explained by lack of unionization and other forms of labor protection

b) The inequality of hourly wages increased, with GINI indexed 1979=100 growing to around 121 in

2010, and the ratio percentile 90/percentile 10 raising from 100 to around 140 (see Figure 6 for wage changes by education and age)

c) The Congressional Budget Office’s household ‘comprehensive income’, which equals pre-tax cash income (wages, salaries, self employment income rents, interest, dividends and capital gains, retirement benefits etc., plus in –kind benefits such as Medicare, Medicaid food stamps, housing assistance, school lunches, energy assistance, etc.), shows that the top 1%, 5% 10% and 20% have all increased their share in the total during 1979-2007, while all the other quintiles have lost grounds, particularly the bottom two which have lost more than 20% of their share, with the remaining two quintiles having lost each around 10% of their share (see Figure 7) As a result, around the crisis year

2007, in the US, the ‘bottom’ 80% of the households had a share of less than one half of the entire society (47.5%) The top 1% fared particularly well, and according to various income definitions, it enjoyed something of a doubling of the share of household income

-15%

-10%

-5%

0%

5%

10%

15%

20%

25%

30%

35%

Age 25-34 Age 35-44 Age 45-54 Age 55-64

High school only Bachelor's only Advanced

Figure 6 Real Hourly Wages by Education and Age, Percentage Variation between 1979 and 1910

Source: Author’s calculation from Table 8.A2 in Jenkins et al (2011), based on Outgoing Rotation Group files of the

Current Population Survey (CPS ORG)

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Top 20%

Fourth 20%

Middle 20%

Second 20%

Bottom 20%

Figure 7 Shares of CBO Household after-Tax ‘Comprehensive Income’, 1979 and 2007

Source: Author’s calculation based on Figure 8.6 in Jenkins et al (2011)

Note: CBO is defined as the Congressional Budget Office's 'comprehensive income' measure: it equals pre-tax cash income plus income from other sources Pre-tax cash income is the sum of wages, salaries, self-employment income, rents, taxable and non-taxable interest, dividends, realized capital gains, cash transfer payments and retirement benefits plus taxes paid by businesses (corporate income taxes and the employer's share of Social Security, Medicare, and federal unemployment insurance payroll taxes) and employees' contributions to 401 (k) retirement plans Other sources

of income include all in-kind benefits (Medicare, Medicaid, employer-paid health insurance premium, food stamps, school lunches and breakfast, housing assistance and energy assistance)

Having established that in the US, and elsewhere, there have been a sustained series of major redistributive shocks, let us turn to the causal mechanism that may have moved the US economy from there to the crisis

As optimistically noticed by Mulligan, the pre-crisis years have been exceptionally positive in terms

of corporate profitability This was true despite the slowdown of GDP growth, which was in excess of 4% per year in real terms in the 1950s and the 1960s, in excess of 3% in the subsequent three decades, and around 2.6% per year 2000-2007 While the growth of both non financial and financial sector profits mostly tracked GDP growth until around 1990, then the former sky-rocketed, thereby driving also the latter

A most revealing aspect of this story is the decoupling of profits and investments, as shares of GDP Since the 1980s non residential private investment has been decreasing, while profits have been increasing, with a paradoxical opposite trend in 2003 In 2006, the year before the crash, the share of recorded profits as percent of GDP was more than four times the non-residential investment In terms of cash flows this must explain the booming incomes of the top income earners, whose income is obviously mostly capital income Growth of profits as share of the GDP is the counterpart of the abnormally high return to capital We need to dissect this point to understand the role of the income distribution shock

When placed in historical perspective, the share of profits out of GDP in the US had been occasionally higher as compared to recent years; but between around 1960 and 1985 the share was mostly decreasing, and then it began to increase strongly, nearly doubling As the share of profits out of GDP changes, something else must accommodate the change From the previous discussion, it seems that the most likely candidate is the share of labor income (in the broadest definition, net of compensation of top managers etc) Under a regressive redistribution shock, you would need to see two partially countervailing effects on savings propensity for any real interest rate The greater share of income going to top income earners would push up the savings rate, while the shrinking of the median household income would push it down Figure 8 shows the personal saving rate in the US, from 1947 to 2010 In the years up to 1982, the saving rate gradually, and with cyclical oscillations, increased from around 4% to nearly 11% by the early 1980s Since then, it collapsed to 1.4% in 2005, and remained at around that level, because of the recent increase due to the precautionary attitude induced by the crisis in the last three years In principle, this effect is compatible