Productivity Growth and the Productivity Growth and the Distribution of Income: Distribution of Income: Results and ExplanationsResults and Explanations

Robert J. Gordon Robert J. Gordon Northwestern University and NBERNorthwestern University and NBER

(co-author Ian Dew-Becker) (co-author Ian Dew-Becker) Speakers’ Series, Ottawa, February 16, 2007Speakers’ Series, Ottawa, February 16, 2007

Department of Finance CanadaDepartment of Finance Canada

22

Ian in SF, you can’t see Ian in SF, you can’t see “MV=PY”“MV=PY”

33

Everyone Knows that US Everyone Knows that US Inequality has increased, what Inequality has increased, what

is new here?is new here? For decades, U. S. data on median For decades, U. S. data on median

family income and median real family income and median real wages show virtually no growthwages show virtually no growth

But U. S. productivity growth has But U. S. productivity growth has exploded since 1995 and especially exploded since 1995 and especially during 2001-04.during 2001-04.

Where did the extra productivity Where did the extra productivity growth go? If the median wage growth go? If the median wage earner didn’t get it, who got it? earner didn’t get it, who got it?

44

The New Elements in Our Data The New Elements in Our Data Analysis and InterpretationAnalysis and Interpretation

In part this presentation is a sequel to our In part this presentation is a sequel to our 2005 BPEA paper, where we were the first 2005 BPEA paper, where we were the first toto– Link NIPA and IRS dataLink NIPA and IRS data– Unravel the puzzles of stable labor’s share, Unravel the puzzles of stable labor’s share,

rising mean wage income, and stagnant rising mean wage income, and stagnant median wage income. median wage income.

Our explanation moves beyond some of Our explanation moves beyond some of the literature bythe literature by– Distinguishing between causes at the bottom Distinguishing between causes at the bottom

(0-90) and at the top (90-99.99)(0-90) and at the top (90-99.99)– At the top, trying to sort out explanations At the top, trying to sort out explanations

involving SBTC, Superstars, and CEO payinvolving SBTC, Superstars, and CEO pay

55

Productivity Growth vs. Median Productivity Growth vs. Median Real Wages and Median Real Real Wages and Median Real

Household IncomeHousehold Income

Labor’s share of domestic income was basically flat Labor’s share of domestic income was basically flat between 1997 and 2005. Implies CPH growth = LP growthbetween 1997 and 2005. Implies CPH growth = LP growth

But…But…– Median wages grew at half the rate of productivity Median wages grew at half the rate of productivity

between 1995 and 2003between 1995 and 2003– Real median family income fell for five straight years Real median family income fell for five straight years

between 1999 and 2004, before rising in 2005. 2005 between 1999 and 2004, before rising in 2005. 2005 was 2.8 percent below 1999 and only 16 percent above was 2.8 percent below 1999 and only 16 percent above 1973.1973.

– Yet 1999-2004 was a period of buoyant productivity Yet 1999-2004 was a period of buoyant productivity growthgrowth

The conflict between mean growth and median growth The conflict between mean growth and median growth poses a basic question: is it a measurement issue or an poses a basic question: is it a measurement issue or an income distribution issue?income distribution issue?

66

Our Headline Result in 2005Our Headline Result in 2005 Over the period 1966-2001 only the top 10 Over the period 1966-2001 only the top 10

percent of the income distribution had real percent of the income distribution had real compensation growth equal to or above the rate compensation growth equal to or above the rate of economy-wide productivity growthof economy-wide productivity growth

Today’s presentationToday’s presentation– Reviews our basic 2005 resultsReviews our basic 2005 results– Updates macro data on productivity trends and labor’s Updates macro data on productivity trends and labor’s

shareshare– Updates Tables 1 and 2 of the 2005 paperUpdates Tables 1 and 2 of the 2005 paper– Provides a more complete review of explanations of Provides a more complete review of explanations of

increased US inequality at the bottom (0-90) and at the increased US inequality at the bottom (0-90) and at the top (90-99.99)top (90-99.99)

– Adds a preliminary review of international dataAdds a preliminary review of international data

77

The Enormous Discrepancy The Enormous Discrepancy Between Productivity Growth Between Productivity Growth

and Real Wage Growthand Real Wage Growth The basic puzzle: as of July 2005, NFPB productivity growth The basic puzzle: as of July 2005, NFPB productivity growth

2001:Q1-2005:Q1 was 3.89 and real AHE only grew at 0.49. 2001:Q1-2005:Q1 was 3.89 and real AHE only grew at 0.49. How can we explain this enormous gap? Was there a How can we explain this enormous gap? Was there a massive shrinkage of labor’s share?massive shrinkage of labor’s share?

Explanation #1: data revisions. 2001-05 productivity Explanation #1: data revisions. 2001-05 productivity growth was reduced from 3.89% to 3.44%. Now in growth was reduced from 3.89% to 3.44%. Now in February 2007 that same number is 3.38%. Extended to February 2007 that same number is 3.38%. Extended to 2006:Q4 is 2.99%.2006:Q4 is 2.99%.

Explanation #2: trend vs. actual. The trend barely reached Explanation #2: trend vs. actual. The trend barely reached 3.0 percent. 3.0 percent.

Explanation #3: Full economy productivity 0.5% slower Explanation #3: Full economy productivity 0.5% slower than NFPB. than NFPB.

Further ExplanationsFurther Explanations– Alternative Wage IndexesAlternative Wage Indexes– Alternative DeflatorsAlternative Deflators

88

8-quarter Actual NFPB Output 8-quarter Actual NFPB Output per Hour vs. the Average Trend per Hour vs. the Average Trend

(through 2006:Q4)(through 2006:Q4)

-2

-1

0

1

2

3

4

5

6

1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005

99

Productivity Growth in the Productivity Growth in the Total and NFPB Economy, Total and NFPB Economy,

1950-20051950-2005

1010

The Macro Data Analysis The Macro Data Analysis Involving Productivity and Involving Productivity and

Compensation Growth, Table 1Compensation Growth, Table 1

This provides data on the entire This provides data on the entire economy, not just the NFPB sector. economy, not just the NFPB sector.

The evolution of productivity growth The evolution of productivity growth compared to compensation growth compared to compensation growth differs greatly by specific historical differs greatly by specific historical intervalinterval

2001-06 retraces the steps of 1997-2001-06 retraces the steps of 1997-20012001

1111

Two Concepts of Labor’s Two Concepts of Labor’s ShareShare

Two ConceptsTwo Concepts– Straightforward share of NIPA employee Straightforward share of NIPA employee

compensation in net domestic factor incomecompensation in net domestic factor income– Add in labor’s part of business proprietors’ Add in labor’s part of business proprietors’

incomeincome Both concepts are expressed as a Both concepts are expressed as a

percentage not of GDP but of domestic percentage not of GDP but of domestic income at factor cost (excludes income at factor cost (excludes depreciation and indirect bus taxes)depreciation and indirect bus taxes)

What to noticeWhat to notice– Up-down cycle 1997-2006 repeats 1987-97Up-down cycle 1997-2006 repeats 1987-97– Share was higher in 70sShare was higher in 70s– Comprehensive concept no change since 50’sComprehensive concept no change since 50’s

1212

What has Happened to What has Happened to Labor’s Share?Labor’s Share?

60

65

70

75

80

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005

Compensation

Compensation with labor component of Proprietor's income

1313

Lack of Connection betweenLack of Connection betweenLabor’s Share and InequalityLabor’s Share and Inequality

Incomes were much more equal in Incomes were much more equal in 1950s but labor’s share was the same 1950s but labor’s share was the same (or lower for the narrow measure)(or lower for the narrow measure)

Much of the rise in inequality > 90Much of the rise in inequality > 90thth percentile occurs in labor income, not percentile occurs in labor income, not capital incomecapital income

The main story is increased skewness The main story is increased skewness within labor income, not a shift from within labor income, not a shift from labor to capital incomelabor to capital income

1414

What is Happening with theWhat is Happening with theNonlabor Share?Nonlabor Share?

FIgure 2b. NIPA Nonlabor Income Share by Component, 1950-2005

0

5

10

15

20

25

30

35

40

   1950      1955      1960      1965      1970      1975      1980      1985      1990      1995      2000      2005   

Perc

en

t/100

Government Enterprises and Transfer Payments

Proprietor's Income

Interest

Corporate Profits

Rent

1515

Some Things to Think AboutSome Things to Think About

Apparent regime change around 1966Apparent regime change around 1966– No good explanation so farNo good explanation so far– Our macro data analysis helps by linking Our macro data analysis helps by linking

labor’s share increase in late 1960s to the labor’s share increase in late 1960s to the productivity growth slowdownproductivity growth slowdown

Share is similar now to 1997. Smoothly Share is similar now to 1997. Smoothly varied in small range for past 30 yearsvaried in small range for past 30 years

So what’s all the fuss about? It’s not that So what’s all the fuss about? It’s not that capital is gaining relative to labor, it’s capital is gaining relative to labor, it’s whowho is getting labor’s shareis getting labor’s share

1616

A Proviso: The DramaticA Proviso: The DramaticNew Work on Intangible CapitalNew Work on Intangible Capital The Authors: Carol Corrado, Charles The Authors: Carol Corrado, Charles

Hulten, and Dan SichelHulten, and Dan Sichel Their Basic Result: About $1 Billion Dollars Their Basic Result: About $1 Billion Dollars

in “Intangible Capital Investment” has been in “Intangible Capital Investment” has been omitted from U. S. National Accountsomitted from U. S. National Accounts

On the Income Side, this is all unmeasured On the Income Side, this is all unmeasured corporate retained earningscorporate retained earnings

Implication: Decline in Labor’s ShareImplication: Decline in Labor’s Share How Much do they Exaggerate its How Much do they Exaggerate its

Importance?Importance?

1717

The Inconsistent Wage Indexes, The Inconsistent Wage Indexes, see Table 2see Table 2

CPH, ECI, and AHE all tell different storiesCPH, ECI, and AHE all tell different stories– AHE only covers production/non-supervisoryAHE only covers production/non-supervisory

ECI is smoother than CPH, but not linked ECI is smoother than CPH, but not linked to NIPA datato NIPA data

Abraham et al. (1999) argue that most of Abraham et al. (1999) argue that most of the AHE-CPH gap is due to AHE’s samplethe AHE-CPH gap is due to AHE’s sample– Production workers not only make less, but Production workers not only make less, but

have less growthhave less growth– AHE vs. compensation reflects the difference AHE vs. compensation reflects the difference

between median growth and mean growthbetween median growth and mean growth

1818

Our Micro Research: Linking Our Micro Research: Linking the IRS and NIPA Datathe IRS and NIPA Data

To whom do the benefits of productivity To whom do the benefits of productivity growth accrue? growth accrue?

Our contribution is a measurement of Our contribution is a measurement of income inequality with a direct comparison income inequality with a direct comparison to productivity growthto productivity growth

Thus we focus on which percentiles of the Thus we focus on which percentiles of the income distribution received real income income distribution received real income gainsgains

We started noting that medians grew much We started noting that medians grew much slower than averages. Here we uncover slower than averages. Here we uncover the nuts and bolts of why this happenedthe nuts and bolts of why this happened

1919

Differences with Differences with Piketty-Saez on U. S. Piketty-Saez on U. S.

We have in common: reliance on tax dataWe have in common: reliance on tax data Their approach: look only at top 10% but Their approach: look only at top 10% but

over a long period (U. S. starting in 1913, over a long period (U. S. starting in 1913, France starting in 1901)France starting in 1901)– Their denominator (total income) is not from IRS Their denominator (total income) is not from IRS

but from national accountsbut from national accounts We look at entire tax distribution from zero We look at entire tax distribution from zero

to 99.99 (not just 90-99.99)to 99.99 (not just 90-99.99)– Our denominator is total reported tax income, Our denominator is total reported tax income,

not national accounts (but we compare the two)not national accounts (but we compare the two) At the end: comments on US vs. Canada, At the end: comments on US vs. Canada,

UK, France, and JapanUK, France, and Japan

2020

Sources of Income Inequality: Sources of Income Inequality: IRS Microfile DataIRS Microfile Data

Cross-sectional data for 1966-2001Cross-sectional data for 1966-2001– Heavily oversamples richHeavily oversamples rich– Allows analysis of top .1% or .01%Allows analysis of top .1% or .01%– 100-200,000 returns per year100-200,000 returns per year– 3,000+ returns in top 0.01 percentile out of 13,000 total 3,000+ returns in top 0.01 percentile out of 13,000 total

filersfilers This study is based on roughly 5 million data This study is based on roughly 5 million data

points, a few more than the typical time series points, a few more than the typical time series quarterly postwar data analysis!quarterly postwar data analysis!

The IRS micro data file provides every type of The IRS micro data file provides every type of income on tax returns – wages & salaries, rent, income on tax returns – wages & salaries, rent, interest, dividends, business income, pensionsinterest, dividends, business income, pensions

~90-95% of tax units file each year~90-95% of tax units file each year

2121

Advantages of IRS Data overAdvantages of IRS Data overCE/CPS Data Used by OthersCE/CPS Data Used by Others

Other papers based on CE/CPS data understate Other papers based on CE/CPS data understate increase in inequalityincrease in inequality– We find half of increase in inequality represented by 90/10 We find half of increase in inequality represented by 90/10

ratio, the other half is within 90-99.99ratio, the other half is within 90-99.99 CE/CPS data are top-coded, e.g., $35,000+ in 1972-CE/CPS data are top-coded, e.g., $35,000+ in 1972-

73 73 Recall bias may vary with incomeRecall bias may vary with income IRS data are linked to actual records, W-2s and IRS data are linked to actual records, W-2s and

1099’s1099’s What do we add?What do we add?

– Adjusting for non-filersAdjusting for non-filers– Eliminating negative nonlabor income Eliminating negative nonlabor income – Adjusting IRS income for fringe benefits and changing hoursAdjusting IRS income for fringe benefits and changing hours

2222

Comparison of IRS 90/10Comparison of IRS 90/10to CPS from Autor-Kearney-to CPS from Autor-Kearney-

KatzKatz

0.9

1.4

1.9

2.4

2.9

3.4

1966 1971 1976 1981 1986 1991 1996 2001

Natu

ral L

og

IRS Data

March CPS Data

Morg Data

2323

Increased Skewness Above 90Increased Skewness Above 90is Missed by CPS Studiesis Missed by CPS Studies

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1966 1971 1976 1981 1986 1991 1996 2001

Ind

ex, L

N=

0 in

1973

90/10

99/10

99.9/10

2424

Shares of New W&S, 1997-Shares of New W&S, 1997-20012001

80-9014.8%

50-8023.4%

90-9511.0%

95-9914.3%

99-99.916.2%

20-5010.8%

99.9-1007.7%

0-201.9%

2525

What About Productivity?What About Productivity?

Adjust W&S upwards as wages take smaller Adjust W&S upwards as wages take smaller share of compensation (~0.4%)share of compensation (~0.4%)– No assumption about level of W&S/Comp, just No assumption about level of W&S/Comp, just

that change is same for everyonethat change is same for everyone Add +0.22% for change in hours per tax Add +0.22% for change in hours per tax

unitunit– Assume changes in hours affect all equallyAssume changes in hours affect all equally

Full economy LP averaged 1.54%, Full economy LP averaged 1.54%, comp/GDP rose from 56% to 59%. Comp comp/GDP rose from 56% to 59%. Comp should follow LPshould follow LP

2626

Almost Nobody Keeps UpAlmost Nobody Keeps Up The headline result: The headline result: only the top 10% have only the top 10% have

experienced adjusted real income gains experienced adjusted real income gains equal to or faster than productivity growthequal to or faster than productivity growth

9090thth percentile grows at 1.77%, 95 percentile grows at 1.77%, 95thth at 2.06% at 2.06% Everybody else slower than 1.54%Everybody else slower than 1.54% Productivity growth has not raised median Productivity growth has not raised median

wages – adjusted growth of median is only wages – adjusted growth of median is only 0.9%0.9%

Could people be moving up across Could people be moving up across percentiles enough to account for this?percentiles enough to account for this?

2727

Adjusted Growth RatesAdjusted Growth RatesAdjusted Percentiles

Year 20 50 80 90 95 99 99.9

1966 7,242 23,667 42,127 52,683 63,367 99,872 220,6531972 8,554 27,059 49,960 63,817 77,094 120,862 270,3201979 8,916 26,402 53,717 69,531 84,790 137,918 342,0091987 8,353 26,562 57,064 76,457 96,591 169,973 517,6441997 8,496 26,436 58,549 82,285 108,012 215,039 692,9552001 9,335 28,559 63,715 90,473 120,630 239,982 806,157

Percent Change 28.9 20.7 51.2 71.7 90.4 140.3 265.4Average Annual Growth Rate 0.73 0.54 1.18 1.55 1.84 2.50 3.70

Hours Adjusted Growth 0.95 0.76 1.40 1.77 2.06 2.72 3.92

88.190.5

83.283.182.683.7

Percent

of CompensationWage Share

Years 20 50 80 90 95 99 99.9

'66-'72 1.89 1.35 1.96 2.31 2.38 2.29 2.50'72-'79 -0.37 -1.32 0.07 0.26 0.39 0.92 2.39'79-'87 -2.45 -1.56 -0.88 -0.45 0.00 0.98 3.55'87-'97 -1.39 -1.61 -1.30 -0.83 -0.44 0.79 1.36'97-'01 0.75 0.33 0.51 0.77 1.16 1.14 2.18Average -0.62 -0.81 -0.17 0.20 0.49 1.15 2.35

Gap Between Productivity and Hours-Adjusted Growth

2828

Labor vs. Nonlabor vs. Total Labor vs. Nonlabor vs. Total Income (Fig 9 in paper)Income (Fig 9 in paper)

Figure 12.Share of Top 10 Percent in Increase of Real Income, $2000, Selected

Intervals, 1966-2001

0

10

20

30

40

50

60

70

1966-79 1979-97 1997-2001 1966-2001

Per

cen

t Labor Income

Nonlabor Income

Total Income

2929

Measurement IssuesMeasurement Issues In 2005 we assumedIn 2005 we assumed

– The change in benefits was the same as the change in The change in benefits was the same as the change in wages in each income quantilewages in each income quantile

– The change in hours of work were flat across the income The change in hours of work were flat across the income distributiondistribution

By limiting our analysis to changes, we did not By limiting our analysis to changes, we did not need to make an assumption about the need to make an assumption about the levellevel relationship between wages and either benefits or relationship between wages and either benefits or hourshours

Benefits increased as a share of compensation, Benefits increased as a share of compensation, from 5 percent in 1952 to 18 percent in 1985. from 5 percent in 1952 to 18 percent in 1985. But flat at 18 percent since 1985.But flat at 18 percent since 1985.– Thus a changing share of benefits to wage and salary Thus a changing share of benefits to wage and salary

income is not an issue in analyzing the increased income is not an issue in analyzing the increased inequality from 1985 to 2005 inequality from 1985 to 2005

3030

How Large is the BiasHow Large is the Biasin our 2005 Analysis of in our 2005 Analysis of

Changes?Changes? Pierce (1999) showed that total comp Pierce (1999) showed that total comp

grew slightly faster than wages at the grew slightly faster than wages at the middle and slower in the tails. middle and slower in the tails.

Compared to our results in his period Compared to our results in his period (1982-96) total comp at the middle (1982-96) total comp at the middle grows 0.2 points faster per year, at the grows 0.2 points faster per year, at the top and bottom 0.4 points slower.top and bottom 0.4 points slower.

No bias in the growth of the 90-10 ratioNo bias in the growth of the 90-10 ratio Limitation: Pierce’s short sample periodLimitation: Pierce’s short sample period

3131

Levels vs. Growth RatesLevels vs. Growth Ratesof Hours by Income Quantileof Hours by Income Quantile

Hours rise with income, as we would Hours rise with income, as we would expect. In 2001:expect. In 2001:– Tax units in 0-20 worked 850 hours per yearTax units in 0-20 worked 850 hours per year– Tax units in 90-100 worked 3850 hours per Tax units in 90-100 worked 3850 hours per

yearyear But we only need information on growth But we only need information on growth

ratesrates What has happened to growth rates of What has happened to growth rates of

hours by income quantile?hours by income quantile?

3232

Growth in comp per hourGrowth in comp per hourWith and without hours With and without hours

adjustmentadjustment The hours adjustment makes little The hours adjustment makes little

difference except at the bottom where difference except at the bottom where hours increasedhours increased

Thus true compensation per hour in the 0-Thus true compensation per hour in the 0-20 quantile fell much more in 1979-97 and 20 quantile fell much more in 1979-97 and rose much less 1997-2001 than in the rose much less 1997-2001 than in the unadjusted IRS dataunadjusted IRS data

Overall, the gap in comp per hour growth Overall, the gap in comp per hour growth rates is slightly smaller between the top rates is slightly smaller between the top and middle, and substantially larger and middle, and substantially larger between the middle and bottombetween the middle and bottom

3333

Original 2005 and Now-Original 2005 and Now-corrected AAGR of corrected AAGR of

Compensation per HourCompensation per Hour

-2

-1

0

1

2

3

4

5

6

7

0-20 20-50 50-80 80-90 90-95 95-99 99-99.9 99.9-100

CPS Hours

Our 2005

3434

Evidence on Evidence on Income MobilityIncome Mobility

While inequality was increasing, there was While inequality was increasing, there was no change in mobility (Bradbury-Katz, no change in mobility (Bradbury-Katz, decade-long transitions within quintiles)decade-long transitions within quintiles)– About 50% in penthouse are still there one About 50% in penthouse are still there one

decade later, same for basementdecade later, same for basement– About 3% make it from basement to penthouse About 3% make it from basement to penthouse

in one decade and vice versain one decade and vice versa– Lots of churning between 20 and 80 Lots of churning between 20 and 80

percentilespercentiles Bottom Line: Increased inequality has not Bottom Line: Increased inequality has not

been offset by increased mobilitybeen offset by increased mobility Opulence of penthouse has increased Opulence of penthouse has increased

relative to basementrelative to basement

3535

Causes of Increased Inequality:Causes of Increased Inequality:Current Debate Based on CPSCurrent Debate Based on CPS

Common Focus on Skill-Biased Technical Common Focus on Skill-Biased Technical Change (SBTC) to Explain 90/50 or Change (SBTC) to Explain 90/50 or 90/1090/10

Since supply of college graduates has Since supply of college graduates has increased, SBTC says that demand must increased, SBTC says that demand must have increased more than supplyhave increased more than supply

Focus on Timing (1980s vs. more Focus on Timing (1980s vs. more gradual process culminating in 1990s)gradual process culminating in 1990s)

3636

PuzzlesPuzzles SBTC Doesn’t ExplainSBTC Doesn’t Explain

– 1989-97 real compensation of CEOs up by 100 1989-97 real compensation of CEOs up by 100 percentpercent

– Real compensation jobs related to computer Real compensation jobs related to computer science increased only 4.8 percentscience increased only 4.8 percent

– Real compensation of engineers declined 1.4 Real compensation of engineers declined 1.4 percentpercent

– Fully half (49%) of income gains in the Fully half (49%) of income gains in the occupational group “managers”occupational group “managers”

– Almost none in occupational groups related to Almost none in occupational groups related to computerscomputers

Why no increase of CEO ratio to average Why no increase of CEO ratio to average worker in Europe, just in U. S.?worker in Europe, just in U. S.?

3737

Income Inequality belowIncome Inequality below9090thth Percentile Percentile

Many articles and hypotheses focus Many articles and hypotheses focus on the timing of changes in the 90-50 on the timing of changes in the 90-50 and 50-10 ratiosand 50-10 ratios

We had previously looked only at data We had previously looked only at data on men and women combinedon men and women combined

But the time path for men and women But the time path for men and women is quite different, and here we present is quite different, and here we present ratios from the latest CPS data (EPI ratios from the latest CPS data (EPI web site)web site)

3838

Ratios 1973-2005 for MenRatios 1973-2005 for MenCPS Ratios for Men Only

-10

0

10

20

30

40

50

1973 1978 1983 1988 1993 1998 2003

50-10

90-50

90-10

3939

Ratios 1973-2005 for Ratios 1973-2005 for WomenWomen

CPS Ratios for Women Only

-10

0

10

20

30

40

50

1973 1978 1983 1988 1993 1998 2003

All5010

All9050

All9010

4040

Organizing Principle for 90-10 Organizing Principle for 90-10 Ratio: Reversal of the Great Ratio: Reversal of the Great

CompressionCompression Elements of the great compression of Elements of the great compression of

the income distribution in 1940-70: the income distribution in 1940-70: rise of unions, disappearance of rise of unions, disappearance of imports and immigrationimports and immigration

Reversal: decline of unions, rise of Reversal: decline of unions, rise of imports and immigrationimports and immigration

Extra elements: equalizing influence Extra elements: equalizing influence of high school educ 1910-40 and min of high school educ 1910-40 and min wagewage

4141

The Role of DeunionizationThe Role of Deunionization

Everyone agrees it mainly affects Everyone agrees it mainly affects menmen

Main source is Card-Lemieux-RiddellMain source is Card-Lemieux-Riddell Main conclusions:Main conclusions:

– Union wage distribution compressedUnion wage distribution compressed– Small effect, just for males, maybe 14 Small effect, just for males, maybe 14

percent of growth in variance of male percent of growth in variance of male wages 1973-2001wages 1973-2001

– SOWA 2006-07 has similar conclusions SOWA 2006-07 has similar conclusions in a different metricin a different metric

4242

Second Aspect of Second Aspect of Great Compression: ImportsGreat Compression: Imports

Trade, Imports, Job DisplacementTrade, Imports, Job Displacement SOWA imply job losses across the income SOWA imply job losses across the income

distributiondistribution– No real impact on the income distributionNo real impact on the income distribution– Perhaps slightly more job losses at the Perhaps slightly more job losses at the

bottombottom Trade has bigger impact on Trade has bigger impact on

manufacturing employment; raises manufacturing employment; raises inequality if lost mfg jobs are above inequality if lost mfg jobs are above average wagesaverage wages

4343

Third Aspect of Great Third Aspect of Great Compression:Compression:ImmigrationImmigration

Fact: Since 1970 triple the flow of Fact: Since 1970 triple the flow of immigrants as ratio of population and immigrants as ratio of population and share of foreign-born workers in the share of foreign-born workers in the labor forcelabor force

Borjas-Katz reduced form approachBorjas-Katz reduced form approach– Lower real wages of domestic workers Lower real wages of domestic workers

by 3% 1980-2000by 3% 1980-2000– Loss reached 9 percent for domestic Loss reached 9 percent for domestic

workers without a HS degreeworkers without a HS degree

4444

Challenge to Borjas-Katz fromChallenge to Borjas-Katz fromOttaviano and Peri (2006)Ottaviano and Peri (2006)

Replace Partial Equilibrium by General Replace Partial Equilibrium by General EquilibriumEquilibrium

When Immigrants arrive, they stimulate capital When Immigrants arrive, they stimulate capital investmentinvestment

Substitution is not general, immigrants compete Substitution is not general, immigrants compete with each otherwith each other– Implication: New immigration drives down wages of Implication: New immigration drives down wages of

existing foreign-born residentsexisting foreign-born residents Thus we may have been asking the wrong Thus we may have been asking the wrong

question, not about the impact on native question, not about the impact on native Americans but on the wages and skills of the Americans but on the wages and skills of the entire population including the immigrants entire population including the immigrants themselvesthemselves

4545

Minimum WageMinimum Wage

Circumstantial EvidenceCircumstantial Evidence Minimum wage hits women harder than Minimum wage hits women harder than

menmen 50-10 ratio for women increased much 50-10 ratio for women increased much

more than for men and increased more than for men and increased permanentlypermanently

It is hard to think of another convincing It is hard to think of another convincing hypothesis than the influence of the hypothesis than the influence of the minimum wage on the 50-10 ratio for minimum wage on the 50-10 ratio for womenwomen

4646

Skill-biased Technical Skill-biased Technical ChangeChange

The gradual increase in 90-50 for both men The gradual increase in 90-50 for both men and women lends plausibility to this and women lends plausibility to this hypothesishypothesis

Our paper disputes some anti-SBTC Our paper disputes some anti-SBTC arguments that are based on timingarguments that are based on timing

We endorse Autor-Katz-Kearney in We endorse Autor-Katz-Kearney in broadening the concept of SBTC to broadening the concept of SBTC to encompass five groups, “nonroutine encompass five groups, “nonroutine interactive” down to “routine manual”interactive” down to “routine manual”

Reason for skepticism: occupational group Reason for skepticism: occupational group data show low wage increases for engineers data show low wage increases for engineers and computer experts, fast for “managers”and computer experts, fast for “managers”

4747

Increased Inequality at the Increased Inequality at the Top,Top,

99.99 vs. 90.0 percentile99.99 vs. 90.0 percentilePrevious distinctions (Kaplan-Rauh):Previous distinctions (Kaplan-Rauh):

trade theories (Hecksher-Ohlin)trade theories (Hecksher-Ohlin)increasing returns to generalists (A-K-increasing returns to generalists (A-K-K)K)stealing theories (Bebchuk et al)stealing theories (Bebchuk et al)social norms (Piketty-Saez)social norms (Piketty-Saez)greater scale (Gabaix and Landier)greater scale (Gabaix and Landier)SBTC (Katz and Murphy)SBTC (Katz and Murphy)Superstars (Rosen)Superstars (Rosen)

4848

In this context, our 2005 paperIn this context, our 2005 paperintroduced the Superstar vs. introduced the Superstar vs.

CEO distinctionCEO distinction

Our critics of 2005 said “superstars Our critics of 2005 said “superstars account for too little” but we account for too little” but we explicitly includedexplicitly included– Entertainment starsEntertainment stars– Sports starsSports stars– LawyersLawyers– By implication textbook authors, By implication textbook authors,

painters, musicians painters, musicians

4949

Inequality at the Top:Inequality at the Top:Superstars and CEOsSuperstars and CEOs

Sherwin Rosen on the “Economics of Sherwin Rosen on the “Economics of Superstars”Superstars”– Steep earnings-talent gradient at the topSteep earnings-talent gradient at the top– ““Hearing a succession of mediocre singers Hearing a succession of mediocre singers

does not add up to a single outstanding does not add up to a single outstanding performance”performance”

Earnings premium of superstars depends Earnings premium of superstars depends on the size of the audienceon the size of the audience– Magnification through technical change: Magnification through technical change:

phonograph, radio, television, cable television, phonograph, radio, television, cable television, CDsCDs

5050

Critique: There Aren’t Critique: There Aren’t Enough SuperstarsEnough Superstars

Entry level to IRS 99.99 percentile in Entry level to IRS 99.99 percentile in 2001 was $3.2 million2001 was $3.2 million– 99.99 percentile accounted for $83 99.99 percentile accounted for $83

billionbillion ForbesForbes magazine “celebrity 100” magazine “celebrity 100”

– Total is $3.1 billion, average $31 millionTotal is $3.1 billion, average $31 million– Many more celebrities not includedMany more celebrities not included

5151

The New “Census” of Sports The New “Census” of Sports StarsStars

2820 athletes in major league 2820 athletes in major league baseball, basketball, footballbaseball, basketball, football

Total income $7 billion, or $2.48 Total income $7 billion, or $2.48 million eachmillion each

Time series on baseball back to 1988Time series on baseball back to 1988– Average increased from $354,000 to Average increased from $354,000 to

$2.1 million$2.1 million– Inflation adjusted increase 8.9 percent Inflation adjusted increase 8.9 percent

compared to 6.0 percent for top 99.99compared to 6.0 percent for top 99.99

5252

Broadening the Concept Broadening the Concept of a Super-starof a Super-star

Superstars include top-paid lawyers, Superstars include top-paid lawyers, doctors, even economists who refuse to doctors, even economists who refuse to leave Harvard when offered megabucks to leave Harvard when offered megabucks to go to Columbiago to Columbia

A few economists make millions by writing A few economists make millions by writing textbookstextbooks

Phenomenon of “continuity”. Wall street Phenomenon of “continuity”. Wall street salaries raise salaries of business school salaries raise salaries of business school finance professors, which in turn raise finance professors, which in turn raise salaries of economics professorssalaries of economics professors

5353

The CEO PhenomenonThe CEO Phenomenon

This is where the real money is in the This is where the real money is in the 99.99 percentile99.99 percentile

1989-2000 CEO compensation 1989-2000 CEO compensation increased 342 percent compared to increased 342 percent compared to 5.8 percent for median hourly wage5.8 percent for median hourly wage– But this hasn’t happened in Europe (UK But this hasn’t happened in Europe (UK

and Canada are in between)and Canada are in between)

5454

Kaplan-Rauh vs. Our 2005 Kaplan-Rauh vs. Our 2005 PaperPaper

The question is how much of the The question is how much of the WAGE WAGE AND SALARY INCOME (W-2) AND SALARY INCOME (W-2) can we find can we find of the top 0.01 percent? (entry level of the top 0.01 percent? (entry level $3m)$3m)

In our 2005 paper we claimed we could In our 2005 paper we claimed we could find about 60 percentfind about 60 percent

Kaplan-Rauh said we were wildly wrongKaplan-Rauh said we were wildly wrong But in our new paper we come up with But in our new paper we come up with

63 percent63 percent

5555

Core of the DifferenceCore of the Difference

First reasonFirst reason– Our simple arithmetic mistakeOur simple arithmetic mistake– Kaplan-Rauh look at actual distribution not Kaplan-Rauh look at actual distribution not

averagesaverages But the second reason is the big oneBut the second reason is the big one

– They look at contribution of executive pay They look at contribution of executive pay to total AGI income including capital to total AGI income including capital incomes, taxable pensions, and capital incomes, taxable pensions, and capital gainsgains

– We just looked at W-2 Wage and Salary We just looked at W-2 Wage and Salary income income

5656

We asked a different questionWe asked a different questionand the right questionand the right question

How much of total W-2 income in the top How much of total W-2 income in the top 0.01 percent is accounted for by top 0.01 percent is accounted for by top corporate executives (1500 * 5)?corporate executives (1500 * 5)?

Answer 20%Answer 20% Adding in all of Kaplan-Rauh’s other Adding in all of Kaplan-Rauh’s other

executives (private firms, lawyers, sports executives (private firms, lawyers, sports and entertainment stars) brings up to 63%and entertainment stars) brings up to 63%

QED: We were right in 2005: superstars QED: We were right in 2005: superstars and CEOs explain the explosion of and CEOs explain the explosion of inequality at the topinequality at the top

5757

Substantive Hypotheses about Substantive Hypotheses about CEOsCEOs

William Shakespeare William Shakespeare (Hamlet, (Hamlet, I, iv)I, iv)::– ““Something is Rotten in the State of Denmark”Something is Rotten in the State of Denmark”

Why distinguish CEOs from Superstars?Why distinguish CEOs from Superstars?– Because they can choose their own salariesBecause they can choose their own salaries– Because they bribe directors compensation Because they bribe directors compensation

committees with perks and stock optionscommittees with perks and stock options– Because they are involved in criminal activity Because they are involved in criminal activity

on a daily basison a daily basis

5858

Bebchuk-Grinstein Study Bebchuk-Grinstein Study (2005)(2005)

1500 Firms1500 Firms– Average $14.3 million for CEOAverage $14.3 million for CEO– Average $6.4 million for top five officers Average $6.4 million for top five officers

(exactly the mean income of 99.99)(exactly the mean income of 99.99)– Total of $48 billion is more than half of Total of $48 billion is more than half of

income in 99.99income in 99.99 Cause? Compensation increased Cause? Compensation increased

76% more than can be explained by 76% more than can be explained by firm size, rate of return, or growth of firm size, rate of return, or growth of rate of returnrate of return

5959

Alternative Theories of CEO Alternative Theories of CEO PayPay

““Arms-Length Bargaining Perspective”Arms-Length Bargaining Perspective”– Supply and DemandSupply and Demand– Stock market boom should have increased CEO Stock market boom should have increased CEO

pay only temporarilypay only temporarily– No increase in alternative occupationsNo increase in alternative occupations

““Managerial Power” PerspectiveManagerial Power” Perspective– Limited only by “outrage constraint”Limited only by “outrage constraint”

““Scratch my Back” Model (The “Lake Scratch my Back” Model (The “Lake Wobegon Effect”)Wobegon Effect”)– Garrison Keillor (U. S. public radio weekly two Garrison Keillor (U. S. public radio weekly two

hours). “Where all the men are strong, all the hours). “Where all the men are strong, all the women are beautiful, and all the children are women are beautiful, and all the children are above average”above average”

6060

The Startling Hypothesis of The Startling Hypothesis of Gabaix-LandierGabaix-Landier

CEO Pay is Proportional to Market CapCEO Pay is Proportional to Market Cap The Elasticity of CEO Pay to Market Cap The Elasticity of CEO Pay to Market Cap

=1.0=1.0 This is True in all Eras and all CountriesThis is True in all Eras and all Countries Any Shortfall of CEO Pay in Europe is due to Any Shortfall of CEO Pay in Europe is due to

Shortfall in Market CapShortfall in Market Cap A frontal attack on those who question the A frontal attack on those who question the

arbitrariness of CEO Pay in the USarbitrariness of CEO Pay in the US– Accounting ScandalsAccounting Scandals– Backdating of Stock OptionsBackdating of Stock Options

6161

Gabaix’s Hypothesis that Gabaix’s Hypothesis that Elasticity of CEO Pay to Market Elasticity of CEO Pay to Market

Cap = 1.0Cap = 1.0Figure 1. 20-Year Rolling Regressions of CEO Compensation on Firm Size

as in Gabaix and Landier's Table II

-1

-0.5

0

0.5

1

1.5

2

2.5

3

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

± 2 S.E. Bands

Coefficient

Note: The x-axis lists the final year of the regression; standard errors reported are robust.

6262

Why Say More?Why Say More?Just Read NewspapersJust Read Newspapers

Nardelli kicked out as CEO of Home Depot Nardelli kicked out as CEO of Home Depot after six years in which stock price declinedafter six years in which stock price declined– Compensation package on the job $240mCompensation package on the job $240m– Golden Parachute $210mGolden Parachute $210m– Maybe some overlap, but who cares?Maybe some overlap, but who cares?

Bebchuk on Steve Jobs and Apple in WSJ Bebchuk on Steve Jobs and Apple in WSJ 01/06/07 (“Inside Jobs”)01/06/07 (“Inside Jobs”)– Massive backdating of optionsMassive backdating of options– Bebchuk paper “Lucky CEOs” this is a massively Bebchuk paper “Lucky CEOs” this is a massively

widespread and pervasive practice. 12% of widespread and pervasive practice. 12% of public firms were involved.public firms were involved.

6363

The International ComparisonThe International ComparisonPuzzlePuzzle

Data based on the share of the top 1% or Data based on the share of the top 1% or 0.1% uniformly show that income 0.1% uniformly show that income inequality in the US grew the most after inequality in the US grew the most after 1970 (US vs. Canada-UK-France-Japan)1970 (US vs. Canada-UK-France-Japan)

Data on CEO pay show much higher ratios Data on CEO pay show much higher ratios of CEO/avg worker in US than anywhere of CEO/avg worker in US than anywhere elseelse

Next slide shows ratios for the top 0.1% Next slide shows ratios for the top 0.1% from 1920 to 1998 (Piketty-Saez and co-from 1920 to 1998 (Piketty-Saez and co-authors)authors)

6464

Income Share of Top 0.1 Income Share of Top 0.1 Percent,Percent,

Five Countries, 1920-1998Five Countries, 1920-1998

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000

U.S.

Canada

U.K.

Japan

France

6565

Piketty-Saez Comments on Piketty-Saez Comments on France vs. U. S.France vs. U. S.

For U. S., most of the decline happened in For U. S., most of the decline happened in four years of WWII, no recovery after warfour years of WWII, no recovery after war– ““Labor market institutions” and “social norms”Labor market institutions” and “social norms”– High income tax rates > 80%High income tax rates > 80%– ““Shift in society’s views on inequality”Shift in society’s views on inequality”

But their graph shows that drop in the U. But their graph shows that drop in the U. S. started in 1937 and continued to 1965S. started in 1937 and continued to 1965

Other countriesOther countries– Canada and UK mimic the US with a partial Canada and UK mimic the US with a partial

elasticityelasticity– Japan and France inequality virtually the same Japan and France inequality virtually the same

now as in 1945now as in 1945

6666

Conclusions and Further Conclusions and Further ResearchResearch

Not just income and wealth are Not just income and wealth are concentrated, but real income concentrated, but real income growthgrowth

Not just true of capital income, also of Not just true of capital income, also of wage and salary incomewage and salary income

80-90% of the wage distribution does 80-90% of the wage distribution does not experience growth near that not experience growth near that implied by productivity growthimplied by productivity growth

6767

Remaining Unanswered Remaining Unanswered Questions,Questions,

Here We Start on Next DraftHere We Start on Next Draft Gabaix-Landier hypothesis about exec pay Gabaix-Landier hypothesis about exec pay

mirroring increases in market capmirroring increases in market cap– Doesn’t work for 1970-2005 in USDoesn’t work for 1970-2005 in US– Works in wrong direction 1940-1970 in USWorks in wrong direction 1940-1970 in US– Hardly works at all EU vs. US in recent yearsHardly works at all EU vs. US in recent years

Who are all these Super-stars and CEOs?Who are all these Super-stars and CEOs?– Kaplan-Rauh make a good start on 99.99 levelKaplan-Rauh make a good start on 99.99 level– Who are they at 99.9 and 99 and 95 and 90?Who are they at 99.9 and 99 and 95 and 90?

Lots of research left to do, starting with Lots of research left to do, starting with the explanation of cross-country the explanation of cross-country differencesdifferences

Let’s start by Let’s start by talking about Canada!talking about Canada!