I’ve said many times: every economic assertion should be preceded by the words “by this measure.” For big economic questions, you need to look at lots of different measures, lots of different way, to get a feel for what’s going on.
This has come home to me as I’ve considered Jared Bernstein’s ongoing takedown of liberal beliefs regarding inequality, marginal propensity to consume (MPC), and economic growth — epitomized in this remarkably weak-kneed December 2013 paper for the (supposedly) liberal Center for American Progress.
Bernstein is a strong voice for progressive policies. But in this paper and widely elsewhere, he repeatedly pooh-poohs the MPC argument.
That argument briefly stated: poorer people spend a higher percentage of their income/wealth each year, so if income and wealth are less concentrated, more widely distributed, there will be more spending.
But Bernstein repeatedly points to certain measures suggesting that things don’t seem to have worked out that way. Even as inequality has increased over the last 35 years, real per-capita consumer spending has continued to rise:
Okay, by these measures (and with y axes tweaked to depict these measures in a certain way), the MPC argument doesn’t look like it’s been borne out.
But even as he puts great weight on this display, he acknowledges that it isn’t very good evidence. You need a counterfactual: Sure, spending went up, but if inequality had remained the same, would it have gone up faster?
He attempts to answer that question with some regression analyses:
How many remote controls do you have for your TV? Do you care to describe the tangle of cables and wires connecting your various components?
If you’re like almost everyone I know, the answers are “at least three, probably five” and “No!”
Have you had a housesitter recently? How long did it take for you to explain which remotes to use, in what order, to watch TV or switch to a DVD or Netflix? Change the channel? Adjust the volume? Did they take notes?
If you’re like everyone I know, the answers are “at least five minutes” and “Hell yes they took notes.”
Nobody these days can walk into a friend’s house and watch TV without detailed instructions. If you’re like everyone I know, even you need to pull out various manuals and flail around at least once every year or two.
I’ve wondered for years: why hasn’t Apple solved this? Imagine a TV that includes everything including a quality sound amp, with one remote and an elegant user interface. The only things plugged in are cable (maybe), ethernet (maybe), and surround speakers (maybe).
Doesn’t that seem like the kind of thing Apple could do brilliantly?
They could be selling tens (hundreds?) of millions of 30-, 40-, 50-, 70-inch TVs at very high tickets. Yeah, the margins on TVs suck, but that was true of cell phones too, wasn’t it? Ditto desktop and laptop computers.
Apple knows how to add value to otherwise-commodity items, and charge serious money for it. People pay the premium, and Apple gets its unheard-of margins, because their elegant integration is worth it.
So why haven’t they done that for TVs?
My answer, in two (or three) words: set-top boxes. That box is the only thing maintaining cable companies’ ever-more-tenuous control over viewers’ content choices — in particular, controlling access to quality feeds of live sports.
The big breakthrough of ITunes wasn’t (just) the user interface. It was the deals that Steve Jobs cut with content providers, making any content from any of the providers simply and intuitively and instantly accessible.
It’s hard to imagine that the cable companies will cede that kind of control to Apple, at least until straits get as dire as they did for the music industry. Unless Tim Cook is a much more brilliant negotiator than I think he is (or than Steve Jobs was), we’re stuck with set-top boxes for the duration. So we’re stuck in component world, with multiple remotes and cables and multiple interacting user interfaces.
If Apple can’t integrate the set-top box and its UI, and the control over content that goes with it, into the TV, they can’t provide a seamless user interface. No seamless interface, no big margins. And the Apple business model is, in two brief words, “high margins.”
So if you’re reading some article among the endless stream of articles over the years about Apple selling TVs, ask yourself one question: what’s with the set-top box? Has Apple managed to cut a deal with the cable companies?
Quoting FloydNorris: “From the fourth quarter of 2004 through the third quarter of 2008, the companies in the S.& P. 500 — generally the largest companies in the country — reported net earnings of $2.4 trillion. They paid $900 billion in dividends, but they also repurchased $1.7 trillion in shares. As a group, shareholders were paid about $200 billion more than their companies earned.”
It just seems wrong. But I haven’t been able to enunciate, in economic terms, exactly why it’s wrong
The “safe-harbor” stock-buyback provisions of Rule 10b-18 of the Securities Exchange Act, passed in 1982, gave C-suite executives carte blanche to extract rents for their own benefit via stock-price manipulation.
This of course gave them the incentive to do so. And they have done so. The rule turned real business managers who “think like owners” into financial prestidigitators who manage their businesses for their own extractive enrichment, not for the good of the business.
Read the whole thing.
One thing that Lazonick doesn’t discuss (but Mason and Norris do) that seems huge to me: interest payments are tax-deductible for corporations. Dividend payments aren’t. This gives them yet another huge incentive to fund their activities through debt rather than equity issuance — and to borrow money for stock buybacks.
Economists almost universally bemoan the mortgage-interest deduction on efficiency grounds (and equality grounds). I really wonder why they don’t vilify all interest deductions, (especially) including the corporate interest deduction. Given the destructive effect on prosperity of the debt-fueled stock-buyback dynamic, it’s arguably even more pernicious.
We should make Rule 10b-18 much more restrictive or repeal it entirely, and we should remove all interest deductions from the tax code.
His crucial (and I think true) point, in my words: you can’t think coherently about inequality — especially wealth inequality — if you don’t think about age. Older households have more wealth, because they’ve had more time to accrue wealth. There’s always gonna be wealth inequality based on that alone. It’s inevitable, and to a greater or lesser extent (warning: normative claim here), that’s as it should be.
I’ve been pondering this dynamic ever since, trying to figure out how to portray it in ways that let us think clearly about it. I’ve searched for presentations and studies, but — perhaps my Google skills need work — I’ve come up with almost nothing. Matt Bruenig’s recent post is a notable exception. Suggestions are very welcome. In any case, kudos and thanks to Scott for planting that seed.
But the anecdotes, surmises, arguments, and thought-experiments in Scott’s post don’t give us much in the way of systematic data and facts. He gives us “data” from his own personal Social Security history, and suggests that to evaluate the inequality dynamic we should drive around the country, “Go into poor people’s houses,” and eyeball their consumption bundles.
Where’s the beef? I find it hard to think about such things without facts, so I’m hoping I can provide some.
The September 4 release of the Fed’s latest (2013) Survey of Consumer Finance data finally prompted me to dig into it. We now have nine triennial SCF samples starting in 1989, encompassing 24 years — most of the period following the Reagan Revolution. What kind of changes and trends have we seen over that quarter century? How did the wealth/age dynamic look in 1989? What does it look like today? How has it changed?
I’m going to concentrate on real (inflation-adjusted) household net worth — assets minus liabilities, things households own minus what they owe others — because:
1. It’s a good measure of prosperity.
2. There are many ways to to think about income (e.g., Does it include capital gains? Unrealized or realized only? What about undistributed corporate earnings? They belong to the shareholder households, right, so should they be imputed to household income? And etc.) Net worth is much more straightforward: assets at market value (with corporate firms’ value imputed to their ultimate household shareholders), minus debt.
I’ll say again some more: every economic assertion should be preceded with the words “by this measure” — different measures tell us different things, explain things differently — so I’ll give you a bunch of different household net worth measures in hopes that together, they tell some kind of coherent (perhaps even causal) story.
I’ll start with simple measures that are hopefully easy to grasp, and then move into more complex measures that my gentle readers may find illuminative.
First: how wealthy is the typical (median) American household, and the typical household in each age group, compared to 1989?
Key to understanding this graph: it’s not showing how individual households have changed — the age groups aren’t moving cohorts — it’s showing how the wealth of age groups has changed. People who are 65 today and have circa $232K (pink line, 2013) were in the 35-44 age group in 1989 and had circa $100K (dark red line, 1989).
The most obvious point: the typical American household is less wealthy today than the typical household in 1989 — down 4% from $85K to $81K. (So much for “the shining city on the hill” and the vaunted promises of trickle-down, supply-side, baby-drowning government, etc. etc. Would this measure look better if the Reagan Revolution had never happened, or never reached the fever pitch it has attained? That is the question, isn’t it? My answer: Yes. Profoundly better.)
Next: The typical elderly household (65+) is richer today than the typical elderly household was in 1989. The typical younger household (under 65, but especially under 55, and especially under 45) is poorer than it would have been in 1989.
(If you’re 65 or 66, please excuse me calling you “elderly.” It’s just a convenient shorthand.)
Another significant feature to note: the lines are more spread out on the right than they are the left. So elderly households are more richer relative to the youngest households than they were in ’89.
You may wish I’d zoom in so you can better see change in the youngest age group, but I’ll do you one better. Here’s a graph that shows changes in median wealth for different age groups:
The typical elderly household is circa 60% richer than they would have been in 1989. All other age groups are poorer than their 1989 equivalents. 35-to-44-year-olds in 2013 are almost 60% poorer than that age group was in 1989 — exactly the inverse of older households. (The dotted line shows the aforementioned 4% decline for all households.)
“We’ve been transferring money to the elderly!” Right? That’s certainly or probably true, but it ignores a key demographic trend: people are living longer. So: 1. They have more time to accumulate wealth, and 2. Their children inherit later in life. And the proportion of people in the higher age groups is larger than it was in the past. How should we think about that, analytically and normatively? Further research needed.
Also worth noting in this graph: before the recent…debacle of 2007–2009, the median wealth of older households grew a lot, while younger households held generally steady. No age group got (much) worse off, while older groups got much better off. Sounds okay, though not stellar. The Great Whatever greatly accentuated that old-young distinction, eviscerating the median wealth of all but the aged while leaving the typical elderly household mostly untouched. (See:Recessions Are Nature’s Way of Keeping the Little Guy Down.)
Next, average (mean) net worth by age group:
Similar, but quite different.
Older households are a lot richer today, by this measure, as they were by the median measure. (Much less true for the 75+ group.) Younger groups didn’t dive post-2007; they’re closer to the same. The 45–54s fall somewhere in between, and All Households are up a reasonable amount.
This is mostly explained by increased (and varying) wealth concentrations at the top — across all households and within each age group — which I’ll look at below. The shape of the wealth distribution has changed. Just to help visualize that, here’s how income distribution has changed.
(The right side of this graph is cut off because the CPS doesn’t break out income levels at the upper end of the spectrum. )
Like the median graph, the mean graph shows a big increased spread going from left to right — older households are generally more richer today relative to younger households. Here’s a look at that change:
There are too many possible interactions going on in this for me to say much. But it seems to be influenced by the rising importance of big incomes at the top of certain groups when it comes to wealth building. Means get pulled around by big concentrations at the top, and that may be what’s happening, for instance, with the 65-74 group — a relatively small number of households in that group (company-owner and CEO-headed?) building extraordinary wealth in recent years. Those over 75 and 55-64, not so much.
All this shows us differences between age groups, now versus then. But what about inequality? Here’s a shot at that:
This is a fairly standard measure of inequality, wealth concentration at the top, here showing that measure within age groups over the years. (You could also compare the top 10% to 50%ers or to the bottom 90%, pull a GINI index, etc. This one’s handy.)
The most pronounced change is in younger groups. In 1989, the average (mean) 35-44 household was 2.6x richer than the typical (median) 35-44 household. In 2013, the ratio is 7.4x. The ratio has grown for every age group except 75+, though not nearly as much as for households under 35. For all households, the ratio has gone up from 3.9x to 6.6x.
And finally, just so you can collect the whole set, here’s the change in Mean:Median ratio over the years:
If you were in the 35-44 age group in 2010, or are in it now, you were, are, in a very competitive winner-take-all group compared even to similar households in 2007 — much less 1989.
All of these measures tell us that wealth inequality/concentration, growing wealth inequality, is a very real thing, even when you take age into account. Matt Bruenig came to a similar conclusion about the current state of age and wealth, presenting it in a graph that you may find more useful if less detailed than mine:
If you’ve read this far, you should read Matt’s whole post. It’s short and sweet.
Wealth inequality is extreme within all age groups. And as I’ve shown, it’s been getting more extreme. This especially for young households — both competing with their peers, and competing with older households.
Those are important conclusions. Wealth inequality data, pace Sumner, is not “nonsense on stilts.” It just requires some plodding, diligent legwork. Isn’t that what professional economists are supposed to do for a living?
Here’s the Excel source file I used (22mb). See Table 4:
Paul Krugman and assorted others have been puzzling at this question recently, one that I’ve been grinding an axe about for some years. For the first time, I think, Krugman’s highlighted the explanation that I keep going on about:
Inflation helps debtors and hurts creditors, deflation does the reverse. And the wealthy are much more likely than workers and the poor to be creditors, to have money in the bank and bonds in their portfolio rather than mortgages and credit-card balances outstanding.
Higher inflation means debtors pay off their loans in less-valuable dollars. So given an outstanding stock of trillions of dollars in fixed-interest loans/bonds, an extra point of inflation should transfer tens of or hundreds of billions of dollars a year in real buying power from creditors to debtors (without a single account transfer happening). Instantly and permanently.
I got some judicious pushback on this thinking in comments from JKH over at Interfluidity, where I had emphasized that rich households own the banks. (The top 20% actual own more like 85% of corporate equity.) While the 20% certainly have debts, they owe them mostly to themselves as bank shareholders, so The-20%-As-Debtors don’t benefit from the cheaper payback caused by inflation; it’s a wash. The 80% do get that benefit (and the 20% suffer), because the 80% owes money, on net, to the 20%.
I tried to simplify the situation with this:
Isolating one component that bank shareholders can be sure of, in a sea of complex and uncertain effects.
Imagine a bank (owned by shareholders, all in the top 20%) that owns a bunch of 30-year fixed-rate loans (borrowed by 80%ers) that won’t expire for 15 years. No ongoing lending by the bank. It just holds these loans and collects interest.
All interest is paid at the end of the year.
The banks’ interest revenues on those loans this year were $1 billion. Expenses were $100 million. $900 million in profit, all distributed as dividends.
Inflation over the next year (and ensuing) is 1%/year.
The bank again makes $900 million in profits, and distributes it to shareholders.
But the 80%er borrowers only pay $891 million in real buying power, and the 20%er shareholders only receive $891 million.
The next year, $882 million in buying power. Etc.
JKH pointed out out again that this only addresses inflation’s effect on the banks’ assets, not their liabilities. Very good point. Here’s an effort to address it, starting with the Balance Sheet account for Financial Business in the International Macroeconomic Accounts. (I averaged the balance sheets for 2003-2012 to get a big overall picture.) Click for larger image.
Now to ask: from the perspective of the 20%er shareholders/bank-owners, what effect will inflation have on these various assets and liabilities?
In many cases it’s quite uncertain (the whole point of the Interfluidity post linked above). They may presume that “Equities and investment fund shares” will go up along with inflation, for instance, so that’ll be a wash. What about “Insurance, pension, and standardized guaranteed schemes”? (I love “schemes” here.) Inflation should help them there, perhaps a lot, but how much and how will those schemes transform over future years and decades? Again, pretty uncertain.
But debt and debt securities are predictable, at least to the extent that they’re fixed-rate loans. Inflation hurts creditors. Period. Let’s just look at those assets and liabilities (click for larger):
Not all of these are fixed-rate, but we can assume that most of them are. (Estimates welcome.)
Here, banks’/20%ers’ assets (what others owe them) exceed their liabilities (what they owe others) by 12 trillion dollars. They’re net creditors. No duh.
To the extent that this stock of outstanding obligations is fixed-rate, an extra point of inflation will decrease 20%ers’ buying power by $120 billion a year.
Now you may suggest that that’s actually small change. It’s roughly 3% of the 20%ers annual income from capital,* and only $6,000 a year for each of the 20 million households in the top 20%. But you can be quite sure that that number is at least one order of magnitude larger for households in the top 1, .1, or .01%.
If the top 1,000 or 10,000 households (who dominate policy) perceive themselves as losing, say, $600,000 a year in real buying power for each point of inflation, are you curious why they hate inflation?
* Rough calculation of 20%ers’ capital income: (National income of $15 trillion – 60% of income going to labor) * Top 20’s 85% share of capital ownership = $5.1 trillion
If you equate wealth and prosperity (not a crazy equation), then one of the best measures of a country’s prosperity is median household net worth. It arguably says a lot more about prosperity than various income measures. It tells you how the typical household (50% have more, 50% have less) is doing.
If lots of households have lots of wealth, that’s pretty much a description of a widely prosperous country — the kind we built in this country in the 20th century — at least until the late 70s/early 80s.
The best estimates we have of this measure come from the Fed’s triennial Survey of Consumer Finances. It started in ’89, and the ’13 report just came out, so we now have twenty-four years of surveys to look at.
Here’s what that quarter century looks like:
Here it is zoomed in on the bottom 60% of Americans:
Top 10%: Up 61%, even with the post-’07 decline. (Top 1% and .1% got richer way faster even than that).
Upper middle class (60–90%): Up 25-32%. Pretty darned good. (Remember, these are inflation-adjusted dollars.) Imagine if we’d all done that!
Middle class (40–60%): Down 18%. Prosperity! Freedom!
Lower middle class (20-40%): Your net worth is down 50% from 1989. It’s been diving since ’95 — almost 20 years.
Lower class (bottom 20%): A huge spike — you’re up 85%! Except: 1. Your holdings are still trivial — $6,100. You’re only two or three months from being on the street. And: 2. You’re down 40% in the last twelve years. Here’s that graph zoomed in:
If you’re in the bottom 60%, here’s how you’re doing compared to the top ten:
I’ll say it again:
Widespread prosperity both causes and is greater prosperity.
We could all (collectively) be much richer right now, even as we could all be much richer.
A nod to income, and how it builds wealth: Say you’re a 50%er. Now imagine that in 2007, you and your friends were making $94,000 a year instead of the $76,000 you actually made. (That’s how much you would have made if wages had gone up with productivity — GDP per hour worked — since the 80s.)
Paul Krugman takes aim today at me (though he doesn’t know me from shinola), and others of my ilk who are at least somewhat obsessed with coming to a coherent definition of “money.”
…people who spend too much time thinking about money in general — specifically, on trying to decode money’s true meaning and find the real, true measure of the money supply; they end up starting to believe that everything in economics hinges on getting that measure right, when in fact almost nothing does.
He’s certainly — obviously — right that defining “money” coherently would not be sufficient to give economists an understanding of how economies work. But I’m here to suggest that it is a necessary condition — that absent such a definition, economists are inevitably fated to wrestle endlessly with incoherent understandings of how economies work.
Economists are like physicists trying to ply their trade without a coherent, agreed-upon definition of “energy.” (That definition is not conceptually simple, but it is coherent and agreed-upon.) Absent that definition, physicists’ discussions would devolve into exactly the kind of unending, unresolvable mare’s nests that are the ubiquitous norm in economics. Cue Truman’s “one-handed economist.”
Without a coherent definition of “money,” it’s impossible to have a coherent discussion — or arguably, even a coherent understanding — of how economies (inevitably, monetary economies) work.
I’ve written repeatedly (you could start here or here) about what I consider to be economists’ central, crippling confusion: even some of the most careful money-thinkers out there (e.g. Isabella Kaminska, J. P. Koenig) frequently confute “money” with “currency-like things.” It’s understandable — we’ve always considered Roman coins to be “money” — but it’s a vernacular understanding that considered carefully, is conceptually incoherent.
I’ll end by again bruiting my preferred definitions of “money,” “financial assets,” and “currency,” and pointing to my many previous posts explaining those definitions’ undeniable virtue (and difficulties):
Money is the exchange value embodied in financial assets.
Financial assets are legal constructs defining claims. The exchange value of those claims is “money.”
Physical currency consists of physical tokens representing balance-sheet credits (claims), so it is actually an extra step removed from being “money.”
So financial assets (even dollar bills), which embody money, cannot “be” money.
These definitions cannot be “true.” I only suggest that they are the most useful, coherent definitions for thinking about economies that I’ve been able to come up with.
To return to Krugman: by this definition, the most useful measure of the money stock (in understanding and exploring how economies work) is not any measure of currency-like things (M1, M2, MB, even divisia measures), but household net worth.
In the big picture, money = household wealth: households’/people’s net stock of claims on all our real stuff (tangible and intangible). Or to be conceptually precise: the money stock = the exchange value of those claims.
Whose preferences would you say are embodied in our current government?
Non-donors as a group are pretty coherent, and seem to give a good representation of what Americans want.
Donors, perceived as an entity, not so much — the group is downright schizophrenic, in particular due to that anomalous bulge at the right. And that 5% or .5% determines what we get — not the 95%. (Money? Pernicious? Feh.)
Now: ask yourself where the self-professed liberals and conservatives that you know land on the left-hand graph.