Are corporate profits actually very high?
The OWS crowd is driven by many issues, including income inequality and the falling share of national income going to labor. I have no reason to question the income numbers, but I also think income is meaningless (as it double counts earnings that are saved), what matters is consumption inequality. I do wonder, however, about the profit data. Is it accurate? Robin Hanson pointed to this interesting post by Jialan Wang:
So according to Benford’s law, accounting statements are getting less and less representative of what’s really going on inside of companies. The major reform that was passed after Enron and other major accounting standards barely made a dent.
.
Next, I looked at Benford’s law for three industries: finance, information technology, and manufacturing. The finance industry showed a huge surge in the deviation from Benford’s from 1981-82, coincident with two major deregulatory acts that sparked the beginnings of that other big mortgage debacle, the Savings and Loan Crisis. The deviation from Benford’s in the finance industry reached a peak in 1988 and then decreased starting in 1993 at the tail end of the S&L fraud wave, not matching its 1988 level until … 2008.
So the accounting profit data is not accurate. In that case, how would we know if corporations have become highly profitable? The obvious place to look is stock prices. The EMH says equity markets can see though dodgy accounting. In addition, for policy issues related to income distribution we care more about the expected future level of profits over a number of years, rather than merely a snapshot of current profits.
In this post Matt Yglesias shows that stock prices have been very volatile, but in recent years there’s no strong upward trend. Oddly, I think the progressive Matt Yglesias is actually too pessimistic about stock prices–a longer time series would have shown an upward trend.
But are stock prices actually the right metric? After all, stocks measure the discounted present value of future profits. What if the discount rate changes?
We’ve seen real interests rates fall sharply in recent years, to exceptionally low levels. Thus future profits are currently being discounted back into stock prices at a very low discount rate. The fact that the S&P is no higher than in the late 1990s suggests that expected corporate profits have fallen very sharply. (Or has risk aversion increased?)
I also wonder about the problem of multinational investments. Are we interested in corporate profits earned by American firms? Or corporate profits earned using American labor? If it’s the latter, we should add in profits from Japanese transplant factories, and subtract out profits that US MNCs are earning in China. Otherwise you’d end up with anomalies like labor earning less than 1% of national income in the Cayman Islands.
Finance isn’t really my area, so I imagine I’ve made many mistakes here. I await clarification by my many excellent commenters.
Tags:
12. November 2011 at 10:58
>>The EMH says equity markets can see though dodgy accounting.>>
EMH has moved from “the price is right” (so that it can see through dodgy accounting) to “it’s hard to beat the market.”
Your discount rate point is accurate. Low rates support high stock prices by increasing the discounted present value of future cash flows (among other things).
12. November 2011 at 11:04
foosion, OK, but even if markets aren’t perfect would you agree that the EMH at least implies that equity prices are more relevant than accounting profits?
12. November 2011 at 11:50
Might be useful to think about investment banks as the extremis case of dodgy accounting.
In that case, how do you know they’re minting it? Salaries and bonuses is how.
12. November 2011 at 11:59
Scott Stock prices do provide useful info about profits. For example, the Dow reached a peak in 1999 as soon as it “figured” out that profits had stopped climbing (and that some companies, like Enron were “misrepresenting” income, something that was “officially” discovered only two years later.
12. November 2011 at 12:42
I don’t understand. Savings are deferred consumption, so if you are worried about consumption inequality, you should worry about income.
12. November 2011 at 12:43
Wish I hadn’t hit submit. Now I think I see your point about double counting, when you save and when you dis-save.
12. November 2011 at 12:57
I don’t think you can use the results to say profits might not be what they say, only that the SEC filing data is not what they say.
Quoting from the link:
“I used a standard set of 43 variables that comprise the basic components of corporate balance sheets and income statements (revenues, expenses, assets, liabilities, etc.).”
So in aggregate there appears to be some deviation, but it may be other variables besides profits.
Another thing this could be showing is anchoring (estimates sometimes do not deviate very far from initial estimates) or targeting: the CEO wants liabilities = x and they achieve x, or at least achieve the first digit of the target.
And on Yglesias’s point, the longer time series would show his point more clearly: since the 90s there has been a stagnation in the S&P 500 when there used to be exponential growth.
12. November 2011 at 13:22
I’m not sure I buy the argument that a deviation from Benford’s Law is necessarily an indication of fraud. The deviation only means that numbers are being arbitrarily chosen rather than measured from natural processes. In recent years, though, accounting has become much more dependent on management estimates, particularly in finance. That doesn’t mean the numbers are fraudulent. It does, however, mean that corporate profits as measured by GAAP aren’t a perfect measurement of the economic return to shareholders.
12. November 2011 at 14:28
> (Or has risk aversion increased?)
Isn’t this one of the explanations for the housing bubble? Banks/investors demanded safe investments and bad loans were synthesized to give them it.
Also, if we accept the EMH that you can’t beat the market but also believe the numbers are lying, the low prices could be due to the market being unable to beat the lying numbers and divine the truth. Not sure what this would imply – that debt would start to be more attractive than equity since it depends less on understanding the numbers? (Actually, isn’t that true? Haven’t bonds done well over the 2000s+?)
12. November 2011 at 14:31
A strong market in finance products that purport to see through accounting challenges has emerged – including scores to measure accounting risk, cash flow analysis, and asset dispositions.
Remember, the EMH fights both ways. Yes, corps have incentives to dissemble, but the market then creates incentives to tell the truth – transparent balance sheets and accounting should be rewarded in the capital markets by lower costs of financing.
In terms of risk aversion, it’s tempting to say that it has increased – but it’s hard to prove that. That is, it’s hard to separate risk aversion from future expectations about the likely states of the world. Classical risk aversion is derived from the curvature of the utility function (the marginal utility of higher income decreases). So is that what’s going on, or are people just becoming more pessimistic about the future?
12. November 2011 at 15:01
Corporate accounting and reporting seem very far from a natural system to me – things driven by quarterly reports, etc.
We must remember that it is entirely legitimate to manage to numbers, and to take deliberate action to force them to be (as well as look) one way or another. Management might take drastic measures to have EPS of 0.01 rather than lose -0.01. That doesn’t mean EPS of 0.01 represents fraud, but it’s certainly not a natural number.
Stock prices have an odd realtionship to the corporations whose ownership they represent. When you buy a share in the secondary market (the normal case), you are giving $0.00 to the entity you are investing in. You are buying a dividend stream, and making a bet (literally) on what other betters will pay for the stock in the future. So stock values are only loosely “estimates of future value of company X” and rather more “bets about what other people will pay for X in the future”. The difference is quite important.
Stock prices could be relatively low because stocks got used as a source of funding to cover obligations, fear over government behavoir, fear that some or many specific companies will be found to be fraudulent in a major way (ongoing examples), and so on. All of these things could drive down prices even among holders who think the future is generally bright.
12. November 2011 at 15:33
Future profits may be expected to be large but discounted at a higher rate than before.
The appropriate discount rate for future profits is the cost of capital not the risk free-rate. Although risk-free rates are low, it isn’t clear the the cost of capital for firms is particularly low. High volatility of equity prices and low IPO activity both suggest that the cost of equity capital is high. BBB corporate spreads are pretty high too (http://research.stlouisfed.org/fred2/series/BAMLC0A4CBBB)
12. November 2011 at 18:05
The original post by Wang on Benford’s law has been updated by the author – I believe the author wants to step back from the original strong assertion of hankypanky to a more nuanced one.
Maybe a better measure of ‘real’ corporate profits is the running multi year average of stock re-purchases to published profits. Presumably a attempt to do this on internal cannibalization rather than actual growth would show up within a 3-5 year period. Or not – depends on the size of the lamprey attached.
12. November 2011 at 18:49
I seem to remember Bob Murphy taking you to the woodshed for saying that income was meaningless. Seems like a very dogmatic thing to do to stick to your views after he drove such a big hole in them.
12. November 2011 at 18:59
Scott wrote:
“After all, stocks measure the discounted present value of future profits. What if the discount rate changes?”
It definitely does. Now who do you like: Campbell-Cochrane’s external habit formation, or Epstein-Zin’s preferences for early resolution of uncertainty?
12. November 2011 at 19:31
There is no support for habit formation or Epstein-Zin in the panel consumption data. The experimental economists cannot seem to find them either.
12. November 2011 at 20:19
That doesn’t surprise me. The frontier of asset pricing is pretty laughable. But it’s a young discipline.
Putting this here for posterity if anyone argues the discount rate is constant:
Shiller, Robert J, 1981.
“Do Stock Prices Move Too Much to be Justified by Subsequent Changes in Dividends?,” American Economic Review, American Economic Association, vol. 71(3), pages 421-36, June.
http://www.math.mcmaster.ca/~grasselli/Shiller81.pdf
12. November 2011 at 21:28
@B: Indeed, this is what Cochrane said last year “we thought 100% of the variation in market dividend yields was due to variation in expected cashflows; now we know 100% is due to variation in discount rates.”
That’s from an earlier draft of his review of the state of knowledge about discount rates. The current version is here:
http://faculty.chicagobooth.edu/john.cochrane/research/papers/discount_rates_jf.pdf
13. November 2011 at 08:08
IMO the best way to measure profitability is: the cash flow statement, particularly operating cashflow, and tangible balance sheet assets or equity (particularly cash). Its hard to fudge those numbers since they are tied back to hard assets and bank balances. Sure, tangible assets are at historical cost, but if operating cashflow is positive the underlying business is making money on a cash basis. There are lots of items on the balance sheet (deferred taxes, derivatives mark-to-market) that mostly has to do with timing. Exxon makes a whopping 14 Bn in cash from operations per quarter: http://finance.yahoo.com/q/cf?s=XOM
BP is similar except for that, you know, multi-billion dollar spill thing.
By that measure, really, its a two-speed world. large corporations are flush with cash, profitable, and have ready access to capital markets to borrow if they need to. Small to medium business are finding loans hard to get (unless you’re the new hotness, like linkedin or groupon) and have to turn to hedge funds or small banks.
are profits “too high?” i dunno -what’s the right amount? IMO large corporations have a competitive advantage right now since credit stanards are too tight (small competitors cannot expand due to poor credit access). Large corporations, with exceptions, are not innovative, more like large bureaucracies, and mostly play defense to protect their position (I’ve worked at a few).
The best way to “bring down” corporate profits is credit growth (credit is largely consumed by growing companies). increased competition! The best way to foster credit growth right now is for the fed to step on the gas.
13. November 2011 at 09:43
DW, Yes, I do agree with the salaries part of inequality. Of course I don’t think income inequality is a problem, only consumption inequality.
Luis, That’s right.
Marcus, Yes, although arguably the problem was also mistaken forecasts about internet business growth.
Jason & FXKLM, You may be right about that study, but it remains an open question whether reported profits are meaningful. That’s why I’m interested in stock prices.
gwern, Good observation–it’s not an area I know much about–I’m open to suggestions.
statsguy, Good observations–I don’t have the answer.
Bryan, I’m not a fan of the “beauty contest” theory of equity prices. After all, the future value must depend on something fundamental–corporations don’t live forever.
OneEyedMan, Maybe, but do we really have good estimates of the corporate cost of capital? Do we know how many bbb bonds will default?
Vince, Thanks, I’m not qualified to judge your suggestion.
John, You said;
“I seem to remember Bob Murphy taking you to the woodshed for saying that income was meaningless. Seems like a very dogmatic thing to do to stick to your views after he drove such a big hole in them.”
He argued that income was useful because it showed how much one could consume without dipping into capital. When I asked why we’d want to know that, I don’t recall he had a good answer. I’ll change my mind when someone tells me why I should care about income.
B, I don’t know enough about either model to comment. My hunch is that economists simply don’t know the answer. Or maybe there is no correct model.
dwb, I certainly agree we need easier money, and also reduced barriers to competition.
13. November 2011 at 11:57
Scott,
NIPA profits are not subject to accounting machinations. They are at peak relative to GDP. There’s your answer.
As for expected profits, consensus forecasts are for higher-than-trend growth.
Why are P/E’s falling? They typically do when real rates are low or below zero. Even so, P/E’s are higher than historical (non-bubble) levels. The wage-adjusted S&P is far above historical levels, even after a ten-year bear market. The market tends to rise on fiscal and monetary intervention; like a sugar-high, when it subsides, the general tendency is down.
13. November 2011 at 14:04
As someone who studies 10k’s and these kinds of issues for a living, I can make a few comments:
* It’s hard to make generalizations, because the composition of the indices changes over time. The S&P 500 today is not the same S&P 500 of ten years ago. Therefore, I find earnings studies to be of limited use, though price and return studies can be of value since they represent aggregate flows.
* Despite the academic love affair with volatility, historical volatility is a very poor measure of risk.
* Large company margins are at historical highs. Several explanations have been offered for this: globalization offers economies of scale(especially for tech & manufacturing); slower economic growth/output gap means less investment; technological changes (i.e., SAP inventory control) has partially disconnected margins from topline fluctuations and increased efficieny; globalization has lowered effective tax rates; or we are at peak margins. I see peak margins in some places, but not everywhere; otherwise I think all of the explanations are valid.
* PE ratios are low relative to historical ranges, especially when interest rates are considered. The debate is: is the “E” is sustainable? has risk gone up? are we in a treasury bond bubble (lowering interest rates too much)? With a pending break-up of the Euro, looming soveriegn defaults, the potential for a hard landing in both Europe and China, I tend to lean towards the “risk” explanation.
* As far as accounting, it is more transparent than 10 years ago (the bubble years were more interesting for foresnic accounting), but not as transparent as it could be. That said, cash flow and balance sheets are extremely strong, though a significant portion of balance sheet cash is locked up in foreign jurisdictions due to US tax law, which explains why companies that are flush with cash are borrowing to buy shares and pay dividends.
13. November 2011 at 19:09
David, But how do we know the NIPA estimates are accurate? And do they include profits earned overseas? And should profits earned overseas be included in income distribution discussions. Are the shares even owned by Americans?
Here’s an example. Suppose a small egalitarian country like Finland had corporations that had huge overseas earnings. Should those figure in estimates of inequality in Finland?
I have an open mind on this, so you may well be right. It just seems to me there are all sorts of open questions here. It’s not clear to me what we are trying to measure, and why.
Dan, Interesting observations, but a couple comments:
1. Corporate tax rates haven’t really fallen in the US, at least like they have in foreign countries.
2. I’m surprised people expect profits to fall. Not saying they are wrong, but most people expect unemployment to fall gradually over the coming decade. If it did would profits fall? I always assumed profits tend to fall during recessions, not expansions. But I haven’t studied that issue, so I may be wrong.
14. November 2011 at 06:56
“The EMH says equity markets can see though dodgy accounting.”
Does it? I thought the EMH stated that the market would successfully and efficiently internalize available information, not that it was a truly magical entity capable of synthesizing information that is not available to market participants.
Indeed, since trading on “material nonpublic information” (a category into which information about dodgy accounting at public firms would seem to fit) is illegal in this country, at least in a broad set of circumstances defined by the imagination of Federal prosecutors, to the extent that the EMH may imply this… the EMH does not describe our equity markets.
14. November 2011 at 12:08
Scott,
1. Corporate tax rates haven’t fallen in the US, but US companies are more global, paying foreign tax rates. They pay the different between US tax rates and foreign tax rates when they repatriate the profits. So, companies don’t repatriate the profits, lowering their effective tax rates. Some companies (i.e., Seagate) don’t pay much in taxes at all because they locate their operations overseas. Estimates are that over $1 trillion in corporate cash is locked up overseas.
2. When the economy is strong, margins are fat and usually P/E ratios contract. This is highlighted with cyclical companies: the rule of thumb is that they are expensive when their P/E ratios are low but cheap when P/E’s are high. People expect profits to fall because S&P 500 margins are above their historical averages (however, when you disaggregate the S&P 500, there is a disconnect with this observation). There is also concern over the pending collapse of the Euro, which will probably hit profits.
14. November 2011 at 18:26
grcridlan, I don’t mean in every case, just on average. I am discussing stock indices, not individual stock prices.
Dan, Why are stocks cheap when their P/E is high? I thought Shiller claimed the opposite
16. November 2011 at 11:56
Scott, for cyclical companies only. Peak earnings imply high E, which lowers the P/E. If you believe in the EMH, then the P will be independent of cyclicality, so peak earnings lower P/E’s. Since I don’t believe in the EMH, in times of peak earnings, the price rises, but not commensurate with the rise of earnings – thus P/E contracts and the price/value ratio rises. This effect is usually tough to ferret out in top-down statistical studies – too many other variables.
Cyclical companies usually trade on P/B or P/Sales.
The S&P 500 has a mix of non-cyclicals (Coca Cola), soft cyclicals (Target), and deep cyclicals (Caterpillar).
I’m not a big fan of Shiller’s analysis, though his data collection is good.
16. November 2011 at 17:06
Dan, I don’t really follow any of that, but then finance isn’t my area.
17. November 2011 at 06:48
Dan Carroll,
How have your investment returns been? By your analysis they should be Buffet-esque.
18. November 2011 at 07:33
My returns have exceeded the S&P by a healthy margin since I started tracking it in 2000, though not every year (I outperform in down years). Unfortunately, I worked for someone else (3 different employers), except for a short stint on my own. It’s amazing how much ideas get diluted when tossed around in a committee and executed by somebody else. Ego plays a big role.
I’d go out on my own, but I’m terrible at marketing and I need the income.