Inequality in the United States: A brief tour

Inequality in the United States: A brief tour

Inequality in the United States: A brief tour of some facts James K. Galbraith Lyndon B. Johnson School of Public Affairs The University of Texas at Austin McCormick Tribune Foundation Conference Series Chicago Federal Reserve Bank April 2, 2008 The University of Texas Inequality Project

http://utip.gov.utexas.edu The Official Story Second, however, there has been, as we know and discussed over the years, a significant opening up of income spreads, largely as a function of technology and of education with the increased premium of college education over high school, and high school over high school dropouts becoming stronger. The whole spread goes right through the basic system. It is a development which I feel uncomfortable with. There is nothing monetary policy can do to address that, and it is outside the scope, so

far as I am concerned, of the issues with which we deal. Alan Greenspan Testimony to Congress March 5, 1997 The Official Story Second, however, there has been, as we know and discussed over the years, a significant opening up of income spreads, largely as a function of technology and of education with the increased premium of college education over high school, and high school over high school dropouts becoming stronger. The whole spread

goes right through the basic system. It is a development which I feel uncomfortable with. There is nothing monetary policy can do to address that, and it is outside the scope, so far as I am concerned, of the issues with which we deal. Alan Greenspan Testimony to Congress March 5, 1997 If technology and trade affect anything, they would affect manufacturing pay

The idea that inequality in the structure of manufacturing pay has increased systematically is a myth. It has risen and fallen. Inequality in manufacturing pay can be measured directly, easily and accurately. It closely tracks the unemployment rate. This measure peaked in the early 1990s and declined sharply as the economy moved toward full employment Inequality in Manufacturing Pay and Unemployment in the U.S. 1953-2005, Monthly Data

.11 .028 .10 .026 .09 .024 .022

Inequality .08 .07 .020 .06 .018 Unemployment

.05 .016 .04 .014 .03 .012

Unemployment rate (left) Inequality of manufacturing pay (Theil index, right) .02 55 60 65 70

75 80 85 90 95 Shaded areas show recessions.

00 .010 05 The best explanation for inequality in manufacturing pay is, it is almost exactly the same thing as unemployment. Looking beyond manufacturing, inequality in pay more generally, including in services, depends mainly on the participation rate. As the proportion of workers in the

population has risen, so has inequality. Overall pay inequality is a combination of two factors: the effect of participation rates and the effect of unemployment rates. Inequality and the participation rate .065 Inequality for 203 sectors participation rate .060

.68 .055 .66 .050 .64 .045

.62 .040 .60 .58 50 55 60 65 70 75 80 85 90 95 source: BLS data and author's calculations 00

Participation rates also determine the famous stagnating median wage Classic argument: **stagnating** median wage Source: CEPR report, April 2007, p.10 But, not for women $35,000 Real median income by gender 2001 Dollars, GDP deflator

$30,000 MALE $25,000 ALL $20,000 $15,000

FEMALE $10,000 $5,000 2004 2001 1998 1995

1992 1989 1986 1983 1980 1977

1974 1971 1968 1965 1962 1959

1956 1953 $0 Between 1965 and 2000, labor force participation increased by nine percent, creating about nine million jobs, or fifteen percent of total job creation. The share of women in the labor force rose eleven percentage points. That of Hispanics rose ten percentage points. That of African-Americans rose three

percentage points. That of white non-Hispanic males fell eighteen percentage points. To be clear, much of this was the consequence of disruptive economic events including especially vast macroeconomic disruptions in the 1970s and 1980s, and institutional change, including the attack on unions. Many older, white, nonHispanic male workers were forced from work. Nevertheless, the transition in the structure of the workforce is an essential component of the rise of measured inequality in the structure of pay.

Thus, when you break out the workforce by race, the stagnation goes away Real median earnings $40,000 ASIAN Full Time 50-52 workweek, year-round, 2001 Dollars, GDP deflator WHITE

$35,000 ALL $30,000 BLACK $25,000 HISPANIC 2004

2002 2000 1998 1996 1994 1992

1990 1988 1986 1984 1982 1980

1978 1976 1974 $20,000 Conclusion: real median incomes rose for all groups in the late 1990s. Full employment is good for median wages. They were stagnant for a period starting around

1971 and ending in 1983 for whites, 1992 for blacks and around 1995 for Hispanics. The stagnation of aggregate median incomes through 1997 is a composition effect. The hourglass phenomenon has much to do with the rising labor force role of women and minorities. And especially with the rising role of new immigrants in the Hispanic workforce. The problem is not whether people start at the bottom. It is whether they end there. This depends very much on how we treat those groups, as they move into jobs previously held by

unionized male, Anglo workers. Inequality in INCOME, on the other hand, has risen substantially. This too can be measured quite precisely, from income tax and other data sources. It is obvious that the explanation for rising income inequality must come from some other source, than rising inequalities in the structure of pay. How about the stock market?

That works fine. U.S. Income Inequality Between Counties 1969 2005 Plotted Against the NASDAQ Composite, with Three Counterfactual Scenarios of Inequality Growth from 1994 2000 9 8.5 0.04 Its the stock market, s&%#*d 8

7.5 0.035 7 0.03 Inequality 6.5 6

0.025 0.02 Piketty-Saez data would give essentially the same answer. 5.5 5

4.5 0.015 4 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 Natural Log of Nasdaq Monthly Close Between-County Income Inequality - Theil's T Statistic (1yr lag) 0.045

If you remove a handful of counties, related to information technology and finance, most of the rise in income inequality in the late 1990s would not have occurred. U.S. Income Inequality Between Counties 1969 2005 Plotted Against the NASDAQ Composite, with Three Counterfactual Scenarios of Inequality Growth from 1994 2000 9 8.5 0.04

8 7.5 0.035 Natural Log of Nasdaq Monthly Close Between-County Income Inequality - Theil's T Statistic (1yr lag) 0.045 Without7 Manhattan

0.03 6.5 6 0.025 Without Silicon Valley Without 5.5 Top 15

5 0.02 4.5 0.015 4 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 Counties with the largest positive changes in Theil Elements 1994 - 2000

Counties with the largest negative changes in Theil Elements 1994 2000 Theil Element County, State Change 1994 - 2000 New York, New York 0.00517211 Santa Clara, California 0.00468738 San Mateo, California 0.00208153

King, Washington 0.00169613 San Francisco, California 0.00148821 Harris, Texas 0.00147724 Middlesex, Massachusetts 0.00131529 Fairfield, Connecticut 0.00099520 Alameda, California 0.00088503

Westchester, New York 0.00086216 Theil Element County, State Change 1994 - 2000 Los Angeles, California -0.00089362 Queens, New York -0.00070519 Honolulu, Hawaii -0.00065515

Broward, Florida -0.00056938 Cuyahoga, Ohio -0.00036473 Kings, New York -0.00034178 Miami-Dade, Florida -0.00032742 San Bernardino, California -0.00031665 Genesee, Michigan -0.00031147

Clark, Nevada -0.00030658 0.025 0.02 0.015 0.01 Manhattan 0.005 0 -0.005

-0.01 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978

1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

2004 Contribution of NY Counties to US Income Theil Index Contribution of New York Counties to U.S. Income Inequality, 1969-2004 New York Putnam Schoharie Fulton Cortland Franklin

Chemung Chautauqua Nassau Dutchess Yates Genesee Rensselaer Allegany Jefferson Niagara

Westchester Saratoga Lewis Tioga Chenango Herkimer Ulster Oneida Suffolk Schenectady Montgomery

Sullivan Onondaga Washington Oswego Queens Rockland Hamilton Seneca Delaware Livingston Clinton

St. Lawrence Bronx Richmond Schuyler Ontario Wyoming Steuben Wayne Orange Kings

Albany Columbia Essex Madison Otsego Tompkins Erie Monroe Warren Greene Orleans

Cayuga Cattaraugus Broome Bar-height is the contribution of the county to the Theil T-Statistic No good jobs for the unskilled? The spiraling crisis in the credit and housing markets has kept [Phil] Gramm in focus, fairly or not. His employer, UBS, revealed yesterday that investment losses tied to the U.S. housing market reached $37 billion over the last six months. For the last three months, UBS posted a $12

billion loss. Gramm, UBS's vice chairman, said yesterday he was "totally unaware" of his bank's massive holdings of securities tied to subprime mortgages, but, he added, "I'm confident we'll recover." Washington Post, April 2. 2008 Per Capita Income Inequality Across US Counties Over Time 1969 2004

Contribution to Inequality between Counties (Components of the Theil T Statistic) Relatively Impoverished Neutral Prosperous (income above national mean) 1969

1970 1971 1972 1973 Nixons Soviet Wheat Deal 1974 1975

1976 1977 1978 1979 Watch The West

1980 1981 The Big Recession 1982 1983 1984 1985

1986 1987 The Oil Bust 1988 1989 1990

1991 1992 1993 1994 1995 1996

1997 The Tech Bubble 1998 1999 2000

2001 2002 Beltway Bubble Cheney Does Wyoming? 2003

2004 It is obvious that monetary policy influences income inequality: any policy that affects the stock market will affect income inequality. Does monetary policy influence pay inequality? The Federal Reserve denies any effect, blaming technological change. Lets test it

The VAR model The VAR model is a very standard model to analyze covariances and causality; our approach is entirely conventional. Like all VAR analysis, it makes no theoretical prediction in advance. Our model features the yield curve, manufacturing pay inequality, unemployment and inflation The yield curve is an attractive, stable measure of monetary policy stance, well established in the

literature. Its also a good predictor of recessions. The yield curve 4 3 2 1 0 -1 -2 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 Source: U.S. Department of Commerce. 30-day T-Bill vs. 10-year bond rate

After accounting for breaks and outliers, and having checked for stability: Inequality 1. The term structure is the most causal force 2. The term structure does affect inequality 3. The term structure is affected by unemployment but not by inflation Dummy regressions on the Taylor rule

TSt f log CPI t log CPI *,U * U t Unemployment target set at 5.5% Inflation target set at 2% Several cases considered: Inflation/Unemployment above or below target Inflation/Unemployment rising or falling

Above and rising or below and falling Dummy regressions done before and after break in 1983. Results for 1969- 1983 Shows the working of the Taylor rule: -Tightening if inflation is above target (but not rising)

-Easing if inflation is below target (but not falling) -But the explanatory power is very low (5% at best) Results for 1984-2006 Contradicts the Taylor rule: -The Fed does NOT react to inflation (rising or

falling, above or below) -The Fed reacts only to low (and falling) unemployment by tightening and inviting recessions. -The explanatory power has improved dramatically. Politics and the Fed

There is a well respected theory of the political cycle (Hibbs 1974, Nordhaus 1975, Tufte 1978, Alesina and Sachs 1988, Greider 1988, Abrams and Iossifov 2006, Hellerstein 2007). To test our version, we define separate dummy variables for the four quarters preceding a presidential election, depending on which party holds the presidency: REPUP and DEMUP

Do those two political dummies affect our the yield curve (and therefore monetary policy)? Tests of a political monetary policy Partial results Conclusions Inequality in pay is a macroeconomic phenomenon. It is strongly influenced by monetary policy, as well as by other policies affecting unemployment and the participation rate.

Inequality in income is largely a financial phenomenon. It is mainly driven by the stock market. Monetary policy appears to be driven mainly by fear of low unemployment, and by political considerations. Reference: UTIP Working Paper No. 42 The Feds Real Reaction Function: Monetary Policy, Inflation, Unemployment, Inequality and Presidential Politics By James K. Galbraith, Olivier Giovannoni and Ann J. Russo July 17, 2007

http://utip.gov.utexas.edu/papers/utip_42.pdf For more information: The University of Texas Inequality Project http://utip.gov.utexas.edu Type Inequality into Google to find us on the Web

Recently Viewed Presentations

  • LAS 60 CANCIONES MÁS INFLUYENTES - WordPress.com

    LAS 60 CANCIONES MÁS INFLUYENTES - WordPress.com

    Su sencillo Stan cuenta una historia sobre un fanático que está obsesionado con Eminem y que le escribe varias cartas sin recibir respuesta. Eminem comienza cantando como Stan, mientras que en la cuarta canta como él mismo respondiéndole. Al final...
  • Menieres Disease Dr. Vishal Sharma Introduction  Described by

    Menieres Disease Dr. Vishal Sharma Introduction Described by

    1. Endolymphatic hydrops rupture of membranous labyrinth potassium rich endolymph mixes with perilymph sustained inactivation of hair cells & neurons of vestibulo-cochlear nerve bathed in perilymph deafness + vertigo + tinnitus. 2. ed . Sympathetic activity ischemia of cochlear &...
  • <Insert Lesson, Module, or Course Title>

    <Insert Lesson, Module, or Course Title>

    Profile APA, MSc. Statistics, MSc. Economics, MA Arabic, MA Islamic Studies Project Manager/ Sr. Financial Consultant Oracle EBS Instructor in OTS/ TE Oracle Certified professional Oracle Certified Support Specialist Salaih Muhammad (ACA) Muhammad Ali MSc Accounting & Finance Functional Consultant...
  • Literacy To use dictionaries to check the meanings

    Literacy To use dictionaries to check the meanings

    Write Boudicca's rallying speech to her tribe. To paint a portrait of Boudicca showing her personality and appearance. Literacy To identify features of a legend in 'Beowulf.' To create a mythical beast for your own legend. To use dialogue between...
  • Religion and Social Change

    Religion and Social Change

    Max Weber - Religion can be a force for change (social and economic). He rejected Marx's 'economic reductionist' model which argues that religion justifies and reproduces wealth and power for certain social groups.
  • Graphs of Linear Systems - Kyrene School District

    Graphs of Linear Systems - Kyrene School District

    The cost of the security services from Super Locks and Fail Safe depends on the number of days the company provides service.. Write an equation that can be used to calculate the cost for "x" number of days for each...
  • CHICKEN LIVER PATE Place in food processor with

    CHICKEN LIVER PATE Place in food processor with

    Place in food processor with soft butter and salt to taste. Place in serving dish and refrigerate. Delicious on whole grain crackers or toast. DESICCATED BEEF LIVER TABLETS The easy way to eat liver! 6 capsules equivalent to one ounce....
  • Analyzing Tourism Informationon Twitter for a Local City

    Analyzing Tourism Informationon Twitter for a Local City

    Naïve Bayes P/N classification model. Accuracy compared with scoring method. Seed words selection discussion. ... Extracted useful features from tweets, developed a tourism information analysis system. Help people to understand and organize significant information of the target city.