
Steven Kaplan on the Inequality and the Top 1% 11/07/2011
What this covers
Steven Kaplan talks with Russ Roberts about who actually makes up the top 1 percent and top 0.1 percent, drawing on his work with Joshua Rauh. Public-company CEOs are a small slice, and their pay rose at about the same pace as that of lawyers, bankers and others. Kaplan reads this as a problem for the corporate governance capture hypothesis. He also warns about the statistics themselves: top-share figures compare different groups of people at two dates, and a falling top share in a downturn does not mean everyone else gained.
The middle of the conversation is about what does drive top pay. Kaplan points to technology and globalization letting talent operate at larger scale, and to incentive structures such as two and twenty. He works through the arithmetic of a manager running 10 billion dollars, and argues that high fees persist partly because cutting them signals weakness. He treats private equity as better aligned than hedge funds and cites its net-of-fee returns against public markets. He considers a cynical oligarchy explanation possible but notes that similar pay patterns appear in London and Asia.
The last stretch turns to the financial crisis and is the most contested. Roberts argues that bailouts from Continental Illinois to LTCM encouraged leverage and a heads-I-win, tails-you-lose mentality. Kaplan answers that Richard Fuld and Jimmy Cayne lost around a billion dollars each on paper, and that panics in 1873, 1907 and 1929 happened without such bailouts. Roberts replies that the executives sold stock along the way. Both end up leaning toward higher capital requirements.
Kaplan argues that the rise in top incomes reflects scalable talent and fee-based incentives in finance, not CEO capture of corporate governance, while Roberts pushes the counter-hypothesis that bailout expectations let finance executives gamble with other people's money.
- Public-company CEOs are only 2 to 6 percent of the top 1 percent, and their pay rose at the same rate as lawyers and bankers.
- Hedge fund and private equity pay (2 and 20) is driven by scaled capital and a market that clears on signals, not price.
- Roberts argues the 1990s bailout pattern created heads-I-win-tails-you-lose incentives, which Kaplan disputes using the leverage and tax evidence and the history of recurring panics.
This asset isn't compiled yet
You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.
Kaplan rejects the claim that privately held partnerships took less risk, citing a late-1920s partnership, reportedly Goldman Sachs, that put half its capital into a leveraged vehicle in 1928 and nearly went bankrupt after the 1929 crash.
“there was a partnership in the late 20s ... they put up 50 percent of the partnerships capital into that investment vehicle ... there was goldman sachs”
After-tax, the bank CEOs ended up with far less than the pre-tax cash-out numbers suggest, because restricted stock and options are taxed as ordinary income at about 50 percent in New York, forcing them to sell stock to pay taxes as it was granted.
“if you get two billion dollars in restricted stock what do you pay on that in taxes a billion ... you have to sell a billion dollars of stock to pay your taxes”
CEOs of non-financial companies are not unusual within the top 1 percent: their pay rose over time at roughly the same pace as other top groups, which is inconsistent with the theory that CEO pay at public companies is the main source of top-1-percent growth.
“the CEOs weren't unusual ... their pay went up but it more or less over time went up like the rest of the top 1% or the top point one percent”
The fee market does not clear on price because a manager who deviates downward on carried interest sends a bad signal about his talent, so high fees persist even though many people would accept lower fees.
“it doesn't clear on price because if you deviate downward particularly on the the 20% of the profits that is a bad signal you can lower your management fee a little bit”
The leveraged loan market for private equity deals performed well through the crisis because lenders learned lessons from the late-1980s and structured deals more responsibly, which is inconsistent with a simple heads-I-win story.
“the leveraged loan market going back to private equity that market actually did just fine ... because they learned their lesson in the late 80s and structured deals in a more responsible way”
Starting in the 1990s, investment banks went public and federal bailouts of large lenders, such as Continental Illinois in 1984, the 1995 Mexican crisis and LTCM, made it more likely that big lenders would be repaid in full, which encouraged leverage and let executives gamble with other people's money under a heads I win, tails you lose structure.
“we were in a period where it became more likely that the federal government bail you out if you lent money to bad investments ... that made it easier for the executives in those businesses to gamble with other people's money rather than their own”
A falling top share in a downturn does not mean the other 99 percent gained: wealth is not a pizza, and in 2007 to 2009 the bottom groups' incomes also fell, just by less than the top's.
“wealth is not a pizza so if somebody gets a bigger share of the pizza doesn't mean you have less to eat ... our incomes also went down there's just went down a lot more”
Top-income statistics compare different snapshot populations at two points in time rather than the same individuals, so a rise in the top 1 percent's share does not mean the same people got richer; panel data that follow the same people often show a different pattern of gains.
“we are typically not talking about the same people so when people say the top 1% of gotten richer that does not mean that somebody in say 1980 necessarily has learned”
Financial crises recur across history, including the panics of 1873, 1907 and 1929 in which partners and investors lost their own money, so the 2007 to 2008 crisis resembles earlier episodes and the heads-I-win explanation does not fully account for it.
“you had the panic of 1873 similar people losing their own money you had nineteen or seven people losing their own money you had 1928 or 1929 ... and then you had 2007-2008”
Assets under management in private equity grew 30 to 40 times between 1986 and 2005, and hedge fund assets grew from tens of billions to about a trillion dollars by 2004, so the money managers at the top grew markedly faster than CEOs.
“the money under management in private equity increase ... from 1986 to 2005 increased by a factor of 30 40 times ... money under management at 35 billion and 91 to the trillion in 2004”
Roberts counters that the CEOs of Bear Stearns and Lehman did not lose their wealth as Kaplan implied: they sold stock along the way, and Lucian Bebchuk's argument is that they took out roughly as much as they lost, because restricted stock and options are taxed as ordinary income at roughly 50% in New York, forcing sales to cover taxes. Kaplan concedes that the answer is not zero, though pre-tax figures may overstate what they kept.
“they took out about as much as they had left ... they're paying taxes of fifty percent on everything they've been granted”
Three forces explain the rise in top pay: information technology lets talent scale across larger markets, globalization lets investors and firms deploy talent worldwide, and incentive structures like 2 and 20 fees reward the most successful managers.
“technology has changed tremendously ... which allows you to scale your talent ... globalization ... call it incentivization where hedge funds”
Public-company CEOs account for only about 2 to 6 percent of the very top, so even if corporate governance is broken, CEO pay cannot be the main driver of top-1-percent income growth.
“we have somewhere between two and six percent of the very top would come from a public company executive”
Parallel pay growth across CEOs, bankers and lawyers suggests they are all responding to similar market forces, which makes the alternative of coordinated corrupt boards implausible.
“why would it be that you know this one group corporate boards and CEOs are corrupt and are overpaying themselves why would they being paid roughly the same as people who are getting paid through arms-length”
The cynical oligarchy explanation holds that top earners have learned to exploit the political system to enrich themselves; Kaplan finds it possible but notes that similar top-pay patterns appear across countries such as London and Asia, which argues for a market for scarce talent.
“they've all figured out how to exploit the political system to enhance their own feather their own nests it's possible ... you do see similar patterns all over the world”
Hedge fund managers do not pay back a share of losses, so the carry structure gives them asymmetric upside; private equity incentives are better aligned because funds net winners and losers over a 10 to 13 year life.
“the private equity fund typically is out there for ten 13 years and you do net winners and losers within a fund so that's different the incentive alignment and private equity is a little bit better”
Private equity returns have beaten public market returns net of fees, suggesting private equity makes companies more efficient, while venture capital was spectacular in the 1990s but weak over the following decade, and venture capital and private equity are good for society while hedge funds are harder to judge.
“private equity actually the the returns they've consistently been better than the returns to the public market and that's net of fees ... venture capital has not been great in the last ten years”
Kaplan rejects the heads-I-win-tails-you-lose theory: investment bank CEOs such as Richard Fuld of Lehman and Jimmy Cayne of Bear Stearns lost roughly a billion dollars each on paper and would not have made the bets if they had foreseen the crisis, so they were surprised rather than gambling.
“Bear Stearns and Lehman ... they each lost about a billion ... they believed in what they were doing and I think they didn't think that you know what happened could possibly happen”
Roberts argues that when investment banks were privately held partnerships investing their own money, they took less risk, so the heads-I-win story does not hold in the old model.
“when these were privately held firms and they were investing their own money they took less risk”
The corporate-governance-is-broken explanation for CEO pay is not fully refuted by the data, since public-company executives account for only about 2 to 6 percent of the very top, but the parallel pay trajectories make it hard to explain why only one group would be overpaying itself.
“why would it be that you know this one group corporate boards and CEOs are corrupt and are overpaying themselves why would they being paid be being paid roughly the same as people”
Hedge fund and private equity pay typically combines a 1 to 2 percent annual management fee on assets with 20 percent of profits, so a manager of 10 billion dollars earning a 50 percent return collects about 1 billion dollars of profit share plus 100 to 200 million in fees.
“one one to two percent is a hundred to two hundred million dollars a year ... twenty percent of the profits ... he gets a billion dollars of the profit share”
Kaplan argues that the same kinds of crisis happened in 1907 and 1929 without bailouts, so the right policy answer is higher capital requirements rather than changes to bailout policy, and both he and Roberts likely agree on that answer.
“the same thing happened in 1929 the same thing happened in 1907 ... in all three cases what with the right answer be you know higher capital requirements”
The market for fund managers works as a market test: managers who are consistently bad cannot raise money again, though randomness means a good manager can still have bad years and a bad one can get lucky.
“there is a market test ... if you're consistently bad you look you don't get to play that's exactly what happens ... obviously there's a role for randomness”
In 2008 every large financial institution was bailed out with creditors repaid in full except Lehman and Wachovia, where the FDIC imposed a haircut on fixed-income creditors, and this policy pattern may increase moral hazard.
“every financial institution that was large was bailed out a hundred cents its creditors were bailed out a hundred cents on the dollar with the exception of Lehman and I think Wachovia”
Evidence that hedge funds consistently outperform the market is weak, so the high pay of hedge fund managers is hard to justify by skill alone.
“the evidence of consistent out performance is actually pretty weak so I think there are you know some people who do that very well”
Fees cannot keep rising indefinitely: 30 percent carry is observed in venture capital, but 40 percent is not, because managers still need a track record and must leave enough returns for investors relative to alternatives.
“you still have to have a track record you still have to leave enough money for your investors relative to other alternatives that they're still willing to pay that premium”
The SEC's decision to let investment banks raise their leverage ratios substantially in the mid-2000s made the 2007 to 2008 crisis worse, though it does not by itself explain the crisis.
“it was exacerbated by the fact that the SEC let them take on more leverage no doubt ... allowed to increase their leverage ratio substantially”
Recessions hurt the top 1 percent's income share: it fell from about 23.5 percent in 2007 to roughly 17.6 percent in 2009, and 2009 inequality was lower than in any year of Bill Clinton's second term.
“in 2009 the top 1% I calculated at seventeen point six ... inequality in 2009 is actually lower than it was during any year of Bill Clinton's second term”
The top 25 hedge fund investors earned over 25 billion dollars in 2009, an average of about 1 billion dollars apiece, even in a weak year for the very top.
“the top 25 hedge fund investors earned over 25 billion a billion and they averaged a billion apiece”
Summing the roughly 8 million dollar average pay of 500 S&P 500 CEOs gives about 4 billion dollars, so the top 25 hedge fund investors earned five to six times the combined pay of all those CEOs.
“eight million times five hundred is four billion and the 25 top hedge fund investors earned twenty five billion so that's five or six times as much”
The top 0.1 percent includes about 3,000 hedge fund investors and about 3,000 private equity and venture capital investors, with fees of similar magnitude, and investment bankers in the same group earned about 28 billion dollars in total, so the three groups fall into roughly the same size buckets.
“we had about 3,000 hedge fund investors in the top point one percent and about 3,000 private equity and venture capital investors ... investment bankers ... twenty eight billion”
The top 1 percent's pre-tax adjusted gross income share rose from a little under 10 percent in the late 1970s and early 1980s to about 23.5 percent in 2007.
“the top 1% earned roughly 10% of the adjusted gross income in the United States ... it peaked in 2007 at roughly exactly 23.5%”
The study misses privately held and closely held business owners and entrepreneurs, doctors, and celebrities, and its celebrity count is likely low because it relies on Forbes-style lists.
“you don't have doctors ... you don't have privately held firms ... we only had a hundred ... making more than 1.4 million which I think is almost definitely on the low side”
Krugman's data show the top 0.1 percent's share rising starting around 1995, which may be a coincidence or a post hoc fallacy rather than evidence of a structural break.
“his data showed the increased the share going to the top one-tenth of one percent starting around 1995 ... could be a coincidence or a post hoc ergo propter hoc analysis”
Top corporate law firm partners increased their pay from about 1982 by roughly the same percentage as CEOs, which supports a common market explanation for top pay rather than CEO-specific capture.
“top corporate lawyers saw their pay go up from say 1982 to date partners at top law firms have increased their pay by about the same amount of CEOs”
Roberts counters that the bank CEOs took about 500 million dollars out after losing about a billion each, so their losses were far from complete, which makes the heads-I-win story more plausible than Kaplan allows.
“they each lost about a billion when their stock values were wiped out leaving them with almost the same amount as their is their paltry nest egg a mere five hundred million”
Ricardo Rebonato, chief risk officer at the Royal Bank of Scotland, wrote a book showing that value-at-risk models are unreliable, yet he could not stop the bank from taking excessive risks, suggesting incentives rather than knowledge were the problem.
“Ricardo ribbon Otto is the chief risk officer of the Royal Bank of Scotland he wrote a beautiful book ... how the valued risk model is of really just you know we don't really understand it”
Membership in the top 1 percent of adjusted gross income required roughly 300,000 to 400,000 dollars in income, with the threshold over 400,000 dollars in 2007 and near the lower end of that range in 2009.
“it's around somewhere between three and four hundred thousand dollars ... in 2007 it was over four hundred thousand”
Paul Krugman argues that much of the gain in the top 1 percent's income share has accrued to the top 0.1 percent, over periods including the 1990s and 2000s.
“Paul Krugman recently blogged ... much of the gain in the income share of the top one percent is actually accruing to the top one-tenth of one percent”
Kaplan's data suggest that highly talented financial people earn similar per-capita pay across hedge funds, private equity, and investment banking, which he argues is what a functioning market for talent would predict because these professionals move between roles.
“you'd expect there if there is a market you'd expect them to do have similar compensation”
The Kaplan and Rauh study can observe only about 15 to 25 percent of the top 1 percent, roughly 20 percent, and they believe a more aggressive estimate would not exceed about 30 percent because they were conservative in their categories.
“we got somewhere between say 15 and 25 percent so let's say 20 we got 20 percent ... I doubt ... we wouldn't get above thirty”
The 1928 top 1 percent pre-tax share of 23.9 percent is the historical peak, but taxes were much lower then, so the pre-tax comparison is not directly comparable to the after-tax picture today.
“it was 23.9% how the good old days ... I think taxes were much lower back in 1928 than they are today”
The mortgage market and collateralized debt obligations were innovations that went wrong in the crisis, and the failures were concentrated there rather than in the leveraged buyout loan market.
“the mortgage market which was I think you know as an innovation gone awry the balrog is Mark in the CDs market which both were innovations that were relatively new”
The same large investment banks that financed the leveraged buyouts also bought the mortgage securities, so lenders to private equity were not too big to fail in a way that would support Roberts's heads-I-win explanation.
“the same large investment banks that were doing the those everyo financing so that's good that's counter evidence I agree with that”
Lower-level employees at investment banks took out less cash before the crisis than top executives, so the people making the risky decisions were often not those who benefited most from the cash-outs.
“as you go lower down in the organization the amount of money people had taken out actually decreases ... it's not necessarily the people at the top”
Kaplan's central interpretive position is that the market for top talent, not corruption of corporate governance, explains rising top incomes, and that his conclusion is partly contested because competitive talent markets may not fully explain why returns differ across groups.
“they're not manipulating the market that's basically the conclusion we came to and I'm sympathetic that conclusion so I want to push back against it”
In the 2008 Forbes 400, there were 27 hedge fund investors, 33 private equity investors and 29 real estate investors, roughly similar numbers across the three groups, which supports the view that talent is paid similarly across these fields.
“I've got 27 hedge fund investors out of 400 I've got 33 private equity investors and I've got Sur 29 real estate investors”
Disclosure rules reveal that Eric Holder earned 3.3 million dollars as a law partner in the year before becoming Attorney General, which puts a government official's prior earnings in the top 0.1 to 0.05 percent band.
“Eric Holder ... three point three million dollars ... he was a point one percent kind of guy in fact that probably put him in the point oh five percent”
Investment bankers' pay fell after the crisis because banking is procyclical: fewer bankers remained and those left earned less when deal volume dropped, while hedge fund and private equity investors continued doing relatively well.
“investment bankers have really been hit by ... the investment bankers that are left are not doing as well particularly since ... they tend to be very Pro cyclical”
Wealth thresholds for the top 1 percent are roughly 5 to 10 million dollars in assets, but wealth is much harder to measure than income, relying on surveys such as those from Edward Wolff at NYU.
“the top 1% looks like somewhere between you know five and ten million dollars according to its assets”
The motivation for the Rauh-Kaplan study was the popular view, from pundits and some academics, that income inequality is driven largely by CEOs of public companies who capture pay through weak boards and broken corporate governance.
“the motivation for that paper came from listening to ... pundits or even some academics say that you know income inequality was largely driven by CEOs”
Hedge fund investors are often wealthy because the fees they earn are large, but the outperformance evidence is unclear, and so the public benefit or fairness of such returns remains a value judgment, not something data can settle.
“whether it's more than enough is a yeah that is uh is a value judgment”
Kaplan declines to decide between the heads-I-win explanation and his own account, saying he does not know whether bailout expectations had a measurable effect on the crisis.
“I just don't think it's big enough ... I don't know I don't know”
Economists largely failed to predict the 2007 to 2009 recession, and the few who did were mainly lucky, while those who lack a clear recovery plan are benefiting most from the crisis.
“economists who didn't most of us did not predict the recession those of us who did I think we're mainly lucky”