Half Full or Half Empty?

by Russ Roberts on February 28, 2008

in Standard of Living

In this post, I showed that net worth per household had risen dramatically in recent years. Of course, I admitted that the average might be skewed by the high end households but that I thought the median was up as well. The Washington Post has a piece today with the data for the median household, Turns out it’s up 35% in real terms (that is, corrected for inflation) between 1989 and 2004. Pretty good news, and the article is generally sympathetic to the surprising possibility that the middle class is thriving.

But the graphic that goes with the article is much gloomier:

Networth

I don’t know if you can read it, (you can click on it and make it bigger) but here’s the first sentence that goes with the graphic: "Between 1989 and 2004, the net worth of the average American family increased by 35%, but household debt more than doubled as more families used debt to finance day-to-day expenses."

Here is another way to describe the data: "Between 1989 and 2004, Americans accumulated more wealth. They decided to  use some of that wealth for increased consumption today."

Notice the chart to the right on productivity and wages. By ignoring benefits, it paints a much gloomier picture than is the reality.

UPDATE: As Randy pointed out in the comments to this post, $55,000 seems like a lot of debt for the median household. Actually the number is not correct. It’s taken from the Federal Reserve Board’s Survey of Consumer Finances. It’s the median debt among families that have debt, about 3/4 of all families. So it is not analogous to the income in the chart. It’s not clear what kind of debt it includes. When I find out, I’ll post on it.

Be Sociable, Share!

Comments

comments

100 comments    Share Share    Print    Email

{ 50 comments }

dave smith February 28, 2008 at 10:39 am

The picture is probabally better than the article reports since it likely includes subtracts student loans from wealth but does not account for the wealth of increased human capital.

spencer February 28, 2008 at 10:43 am

The difference between average hourly earnings and compensation are driven by two things. One, is the value of fringe benefits. The other is growing inequality. Average hourly earnings measures the wages the some 80% of the population that punch a time clock earn. But the wage portion of the compensation includes all forms of payment, including the stock options that CEOs earn. Fringe benefits are driven by four things. One is social security contributions. since SS taxes were raised in the 1980s this has grown slower then wages because of the cap on SS payments. the second is employee contribution for private retirement funds.
Again this is strongly influenced by the growing inequality of pay and the shift from defined benefits to defined contribution retirement packages which reduced the payments by corporations. the other two factors are Medicare taxes and health insurance. Like, ss taxes the medicare taxes grow with wages. the health insurance payments reflect both real gains and higher costs in the health care system. It is standard practice to deflate these four variables by the overall CPI, the same as wages. for the first three variables this is a reasonable and valid adjustment. But since price increases in health care is higher then the overall CPI deflating the health insurance component of compensation by the overall CPI generates a bias that overstates the growth in real fringe benefits.

Given that the differences in fringe benefits and average hourly earnings are driven by growing wage inequality and higher prices for health care why is it painting a much gloomier picture than is the reality?

From 1950 to 1980 the trend growth rate of real median family income was 2.3%. From 1975 to 2005 the trend growth rate was 1.3%.
If the 1950 to 1980 trend had continued real median family income would now be 50% higher then it actually was. Most of the slower growth since 1975 can be explained by slower productivity growth, but a significant part of the remainder stems from the drop in the US terms of trade.

I will be glad to debate the real issues with you of why growth has slowed sharply over the last 20-30 years, but the strawman you keep pushing that liberals claim there has been no growth does not deserve a rebuttal.

Randy February 28, 2008 at 10:47 am

Question; Is there such a thing as "median productivity"? That is, if the productivity gains are being made by a few technical specialists and global entrepreneurs, why would we expect to see median income rise at the same rate as overall productivity? It seems to me that median income would rise at the same rate as median productivity.

John Dewey February 28, 2008 at 10:49 am

Spencer: "the strawman you keep pushing that liberals claim there has been no growth does not deserve a rebuttal."

Spencer, I don't see how that is a strawman. I've read the claim of flat median income repeatedly in blogs and in essays of liberal writers.

John Dewey February 28, 2008 at 11:00 am

This table bothers me. It would seem to create a profile of the typical American household. But is it? Would the same median income household also be the median debt household and the median retirement account household and the median home value household?

A wealty retiree household could easily have the median income level while at the same time be at the 90th percentile in total assets. Young professionals may have relatively high incomes but median debt levels if still paying off education loans.

I realize that each line in the table provdes useful information, but perhaps grouping all of these together masks the real household economic situation. Not sure I have a better alternative, or at least not a concise one.

Randy February 28, 2008 at 11:06 am

Being quite close to a median income earner myself, I find that $55 thousand non-mortgage debt figure incredible. I have never in my life carried more than $20k in non-mortgage debt, and that includes car payments. I can see how a few people with medical problems might hit the $55k mark, but this is supposed to be a median figure. So are the numbers wrong? – or have people really gotten that stupid?

John Dewey February 28, 2008 at 11:07 am

The caption accompanying the table bothers me even more:

"the net worth of the average American family increased by 35% …"

and then, of course, the table presents the median net worth. I realize that the "net worth of the average family" is not the same thing as the "average net worth of American families". But any table that uses both the terms "average" and "median" is bound to confuse.

FreedomLover February 28, 2008 at 12:18 pm

One could argue that most people are doing WORSE off because they've taken on more debts then ever before. Whether that's their fault or not is up for debate. Another factor is that median wages will be declining in real terms as more jobs get offshored to India.

FreedomLover February 28, 2008 at 12:19 pm

Then there's the reality of student loans in the 20-30K range that never get paid off. It's like debt slavery.

Methinks February 28, 2008 at 12:24 pm

"Between 1989 and 2004, Americans accumulated more wealth. They decided to use some of that wealth for increased consumption today."

Therein lies the problem, according to leftists. They believe that it is the government's job to decide how much wealth should be used for consumption and how much for investment, not individual Americans. Individuals acting in their own self-interest – and worse, deciding for themselves what that self interest is – creates a "problem" which can only be "fixed" with government central planning.

Martin Brock February 28, 2008 at 12:25 pm

Here is another way to describe the data: "Between 1989 and 2004, Americans accumulated more wealth. They decided to use some of that wealth for increased consumption today."

Here's another way: "Between 1989 and 2004, the price of homes and corporate shares rose, and Americans borrowed a lot more."

Rising asset prices and debt are what I expect in an Austrian-style boom driven by cheap credit. "Wealth" can appear to rise faster than debt, because the increase in debt measures the rate at which new money enters the competition to bid up prices, but the rise in "wealth" applies the new, higher prices to everything.

If I give lots of new money to five percent of the population, they can easily bid up prices by fifty percent before the money runs out. They don't accumulate everything or even fifty percent of everything in the process. They accumulate less and less as they pay more and more for it. The other 95% of the population then perceives a great increase in their wealth, but only prices in a narrow category have changed.

I don't imagine any zero sum game here. Real growth is a separate issue. This inflationary process can coincide with real growth.

By ignoring benefits, it paints a much gloomier picture than is the reality.

So what is the reality? The figure on the right shows a decline since 2000. Were benefits not rising before 2000? Have they risen faster since 2000? Are you talking primarily about health insurance premiums?

Randy February 28, 2008 at 12:27 pm

Freedom Lover,

The point about student debt would answer my question above. But doesn't anyone work their way through school anymore?

John Dewey February 28, 2008 at 12:34 pm

freedomlover: "It's like debt slavery"

I don't understand. How is fulfilling an voluntary contractual obligation slavery? Is it debt slavery that I must pay my mortgage debt? Is it debt slavery that I must honor the terms of my business lease? I think a better term to describe paying off student loans is "growing up".

Martin Brock February 28, 2008 at 12:40 pm

They believe that it is the government's job to decide how much wealth should be used for consumption and how much for investment, not individual Americans.

The Austrian view concerns the amount of money flowing toward investment vs. the amount flowing toward consumption. If the "wealth effect" theory is true, we've possibly seen a feedback effect of a monetary expansion driving up the price of rent-generating assets like homes and corporate shares to unsustainable levels that don't reflect real growth. If so, either prices rise more generally (inflation) or the price of rent-generating assets falls.

Brad Warbiany February 28, 2008 at 12:40 pm

I find it interesting the way they say that:

"the net worth of the average American family increased by 35%, but household debt more than doubled"

When you read that, it seems like the doubling of debt is supposed to overshadow that 35% increase in worth, but that's not how it works.

If net worth increased by 35% while debt doubled, that means that total assets more than doubled to make the NET worth increase.

I'm wondering if the writer who added that to the graph did this deliberately in order to craft a certain perception, or whether they were confused by it themselves.

John Dewey February 28, 2008 at 12:52 pm

Brad Warbiany: "If net worth increased by 35% while debt doubled, that means that total assets more than doubled to make the NET worth increase."

I would agree, if the median net worth household was also the median debt household. As I pointed out earlier, the figures shown are all median numbers and have nothing to do with an "average" American family (whatever that is).

brad: "or whether they were confused by it themselves."

My bet is that the writer of the piece was a journalism major who never had to learn either elementary statistics or algebra.

Martin Brock February 28, 2008 at 12:55 pm

This table bothers me. It would seem to create a profile of the typical American household. But is it? Would the same median income household also be the median debt household and the median retirement account household and the median home value household?

That's a good point. The net worth of a household with the median income is not the median net worth. I expect the net worth of a household with the median income to be higher than the median net worth of households, because many retirees are in the middle of the income distribution.

Chris February 28, 2008 at 12:56 pm

Randy –

Unfortunately, a lot of people are addicted to new cars — they buy new ones every few years and roll the old debt into the new loan. It's not terribly unusual for people to have higher car payments than house payments.

The other thing which has risen dramatically is the cost of higher education and, thus the amount of student loan debt. When I graduated in 1990, tuition was around $16K. Today, it's almost $40K, which is about a 5.5% annual increase. That can add a lot of debt in a few years. [Considering that universities are generally non-profit and thus need to charge somewhere close to their costs, I'd be interested to hear why the increase is so much greater than the rate of inflation over the same time.]

Martin Brock February 28, 2008 at 1:03 pm

If net worth increased by 35% while debt doubled, that means that total assets more than doubled to make the NET worth increase.

The problem is that rising price is easily confused with rising wealth. The price of rent-generating assets need not rise at the same rate as the price of consumption goods. I don't expect these prices always to rise together, but if credit is too easy, I do expect prices to rise somewhere.

muirgeomuirgeo February 28, 2008 at 1:35 pm

Just a quick question here as it was claimed that this WAS NOT an economy built on debt….Is the $9,000,000,000,000 federal debt included in these numbers?

If not shouldn't it be?

How about the current massive over estimation of asset values from home values to securities and commodities prepared up by the feds continued interest rate drops and devaluation of our dollar.

If you are claiming this is the grandiose results of free market economies then be prepared for some major backlash against such ideologies.

I think you do free market liberal economics a disservice when claiming that is what we have as opposed to a rigged flawed banking/financial system.

spencer February 28, 2008 at 1:47 pm

happy jugler — you pay attention to the facts, so I would suggest you go over to angry bear and look at a different perspective then Russ Roberts panglossian gloss.

Python February 28, 2008 at 1:54 pm

It must be nice to be able to take 4 year old data, put up a garbage caption title, make a pretty graphic, and be published in a leading newspaper.

I hope I continue to be "worse" off by having my Net Value increase.

Randy February 28, 2008 at 1:56 pm

Muirgeo,

The Federal debt; Is only a real debt if it will someday have to be paid off – and that isn't the case. The only thing that has to be paid is the interest, and that is already covered by the budget.

Over estimation of asset values; I thought of that too, but the numbers are from 2004, so they're probably fairly accurate today.

muirgeo February 28, 2008 at 2:14 pm

I like how wages paralleled productivity during the mid and latter 90's. That's how it should always be. Anything else is just a matter of the money manipulators doing all sorts of financial trickery using the government and the finance laws to increase their asset values being sure to leave the worker and taxpayer holding the bag when the bubble burst. Its bull crap and there is no excuse for claiming this is the way things are and the average Joe should just sit down and shut-up and be thankful for what they have.

I'm sick of the financiers and bankers and money manipulators raving about free markets during the boom and then crying to Uncle Sam in the most socialistic ways for help during the bust.

RANT OUT!

Martin Brock February 28, 2008 at 2:16 pm

Over estimation of asset values; I thought of that too, but the numbers are from 2004, so they're probably fairly accurate today.

But debt has probably continued to rise. The saving rate has fallen since 2004. I assume the BLS conducts this survey every year. I'd like to see the results for 05-07.

Python February 28, 2008 at 2:26 pm

Much of the data shown in the article is taken from the Survey of Consumer Finances that is done every 3 years. The 2007 issue is not complete.

http://www.federalreserve.gov/PUBS/oss/oss2/2004/scf2004home.html#summary

And summarized in:

http://www.federalreserve.gov/PUBS/oss/oss2/2004/bull0206.pdf

Randy February 28, 2008 at 2:30 pm

Muirgeo,

See my question concerning the difference between productivity and median productivity. The point being that I see no reason why anyone should expect rewards for productivity increases they had nothing to do with.

Randy February 28, 2008 at 2:32 pm

Martin,

I hear you, but people running up debt is their own damn fault. I mean seriously, who the hell carries a $55k debt on a $43k salary. That's just stupid. Always has been and always will be.

Brad Warbiany February 28, 2008 at 2:33 pm

Martin,

Nothing that I said is contradictory to the Austrian credit bubble model, I'm simply pointing out that the language they use makes one think that debt is rising faster than assets, whereas it's simply rising faster than *NET* worth. Assets must be rising faster than debt for net worth to be increasing.

Whether those asset prices are rising due to inflationary government credit policies is a different matter, and one I don't dispute.

Methinks February 28, 2008 at 2:35 pm

Freedomlover:One could argue that most people are doing WORSE off because they've taken on more debts then ever before.

Only if the return on the their assets is lower than the interest they're paying on their debt. Since net worth is growing, I'm inclined to believe this isn't the case. So, they are better off taking on more debt.

For example, even though I had the cash to buy my car, I chose to finance the car and invest the cash in my business because the interest rate I pay on the car is much lower than the return on my business. I have more debt, but I'm better off.

rpl February 28, 2008 at 2:39 pm

To compare paying off a student loan to debt slavery requires a great deal of ignorance about the institution of slavery. The salient feature of debt slavery is that the debtor is required not merely to pay off his loan, but to do so by working for whomever purchases his contract at whatever (non-market) wages the employer sees fit to give him. Moreover, debt slaves normally were dependent on their employers to supply food, shelter, and the other necessities of life, again at non-market prices set by the employer's whim. The usual result is that the debt slave becomes more, not less, indebted with time.

About the only feature that a student loan shares with debt slavery is the existence of a debt. The debtor can work for market wages for whomever will employ him, and he buys his food and shelter at market rates as well. Over time the debt decreases, not increases, until the debt is eventually paid. In other words, there is no resemblance at all to anything that could reasonably be called "slavery".

Randy February 28, 2008 at 2:54 pm

Methinks,

True. There's debt and there's leverage. But these are median numbers. How much of it do you suppose is leverage? – and how much is shiny new SUVs?

John Dewey February 28, 2008 at 3:05 pm

Randy: "The point being that I see no reason why anyone should expect rewards for productivity increases they had nothing to do with. "

I agree. But even further, I see no reason that a productivity increase by a specific group of U.S. workers should automatically provide them with a wage increase. U.S. industry has faced significant foreign competition for about 35 years. The increased skills of a billion or more foreign workers now compete directly with U.S. labor. Though the productivity of U.S. workers has no doubt increased markedly, the gap between average global productivity and U.S. productivity has likely shrunk. Wouldn't the increased supply of workers in an industry inhibit the bidding up of wages that might otherwise occur with productivity increases? I'm not an economist, but I think the theory behind productivity and wage correlation assumes a constant supply of labor.

I suspect that if we could compare global productivity changes with global wage changes, we'd see a better correlation.

Methinks February 28, 2008 at 3:26 pm

Randy,

True, you're not earning a return on the SUV. But if you're choosing to invest the cash and finance the purchase because the spread between the return on the invested cash and the interest rate you're paying on the purchase is positive, then financing the purchase is effectively leverage.

If people are simply financing purchases because they don't have the cash, then their net worth will be negative. But the graph clearly shows that this is not the case.

I'm oversimplifying – if you finance a car because you needed one to get to work and couldn't work at all or couldn't earn as much if you didn't have one, there's an implied return on the vehicle which a calculation like this wouldn't account for.

Martin Brock February 28, 2008 at 3:39 pm

I hear you, but people running up debt is their own damn fault. I mean seriously, who the hell carries a $55k debt on a $43k salary. That's just stupid. Always has been and always will be.

I worry less about whose fault it is or how stupid it is than about how widespread it is and how much of recent "growth" is attributable to it. If we're producing a lot of stuff that people can't really afford, that's a problem. The reorganization can be painful.

Jeff S. February 28, 2008 at 3:39 pm

Martin,

You wrote: "because the increase in debt measures the rate at which new money enters the competition to bid up prices, but the rise in "wealth" applies the new, higher prices to everything." Wouldn't this always be the case, whether or not "…(you) give lots of new money to 5 percent of the population…" or you give "x" amount of money to whatever percentage of the population. I'm trying to understand if this is your way of saying that because of easy credit asset prices in the table are inflated and will drop.

Also, you wrote: "The other 95% of the population then perceives a great increase in their wealth, but only prices in a narrow category have changed." Again, isn't this always the case, regardless of what percentage of the population availed themselves of easy credit? I didn't buy any new homes, but speculators on my street drove my home price up. I didn't buy any new stock, but speculators drove up the price of the stocks I hold.

I'm trying to see how your entire post is anything other than just a long way of saying that easy, cheap credit inflated asset values and the table is therefore wrong. Not that I disagree with you.

Randy February 28, 2008 at 3:46 pm

Martin,

"The reorganization can be painful."

True. But if there's going to be a reorganization then placing the blame correctly matters. Stupidity should be painful.

Jeff S. February 28, 2008 at 3:46 pm

"I think you do free market liberal economics a disservice when claiming that is what we have as opposed to a rigged flawed banking/financial system."

"Anything else is just a matter of the money manipulators doing all sorts of financial trickery using the government and the finance laws to increase their asset values being sure to leave the worker and taxpayer holding the bag when the bubble burst."

"I'm sick of the financiers and bankers and money manipulators raving about free markets during the boom and then crying to Uncle Sam in the most socialistic ways for help during the bust."

George,

Your obsession with the "rigged" financial system, "money manipulators" and "financial trickery" indicates you may have descended into a weird, Americanized, Gnomes of Zurich derangement.

There are, for example, probably two dozen coherent arguments against widening income inequality on varied grounds, yet you can't muster a single one?

Martin Brock February 28, 2008 at 3:50 pm

Only if the return on the their assets is lower than the interest they're paying on their debt. Since net worth is growing, I'm inclined to believe this isn't the case. So, they are better off taking on more debt.

This psychology is precisely how easy credit becomes a bubble. More to the point, the borrowing we've seen recently seems to fuel consumption. If it's fueling the consumption of nondurable goods, that's very different than borrowing to buy productive means.

I expect many small businesses to starting going on the market around this time, and that'll show up as rising debt and could also account for a lower saving rate, but that's not how I'm hearing it described.

Methinks February 28, 2008 at 4:06 pm

This psychology is precisely how easy credit becomes a bubble.

This is basic Finance 101.

FreedomLover February 28, 2008 at 4:10 pm

Freedom Lover,

The point about student debt would answer my question above. But doesn't anyone work their way through school anymore?

Posted by: Randy | Feb 28, 2008 12:27:01 PM

My parents paid for my education through community colleges and state universities. Total cost about $8000. But I guess I'm not a real man, because I didn't (a) go to a prestigious university and (b) didn't work my own way through

FreedomLover February 28, 2008 at 4:11 pm

For example, even though I had the cash to buy my car, I chose to finance the car and invest the cash in my business because the interest rate I pay on the car is much lower than the return on my business. I have more debt, but I'm better off.

Posted by: Methinks | Feb 28, 2008 2:35:25 PM

That won't work for the average person who doesn't have high-return investments.

Methinks February 28, 2008 at 4:20 pm

That won't work for the average person who doesn't have high-return investments.

Well, they don't have to have a "high-return". The return on investmentjust has to be higher than the interest they're paying on debt. It's a basic cost/benefit analysis. But I think you're missing my point.

There will always be knuckle-heads who think that if they can make the monthly credit card payment on their shiny new Ducati motorcycle, they can afford it. However, since the average household's net worth is positive and growing, can you really conclude that the average American is making the Ducati knuckle-head decision?

Randy February 28, 2008 at 4:24 pm

FreedomLover,

I'm doing exactly that for my daughters. My point is that working one's way through school, and/or going to a school one can afford, would seem to be preferred options to running up a significant debt. As Methinks points out, there are situations in which running up a significant debt, for school or otherwise, can be a case of leverage, but given the $55k debt to $43k salary ratio, I strongly suspect that leverage is not the explanation.

Randy February 28, 2008 at 4:29 pm

Methinks,

Yes, I suspect there are a lot of Ducatti knuckle heads. Then again, I reckon that the bottom line on the net worth figures is that times are so good that it doesn't really matter.

Russ Roberts February 28, 2008 at 4:32 pm

Randy,

You were right about that $55,000 figure. I have updated the post.

Methinks February 28, 2008 at 4:39 pm

Randy,

I wonder how much of that net worth is real estate which is teetering on either a price cliff or long slope. Incidentally, I worked full time and went through school and took as little debt as possible (although, I was very tired). I paid off the debt in two years. But doesn't education – especially bachelors and masters degrees – have a very high return, making it worth the debt?

Methinks February 28, 2008 at 4:41 pm

"and went through school" = "all through school"

Methinks February 28, 2008 at 4:41 pm

"and went through school" = "all through school"

T. Fry February 28, 2008 at 4:46 pm

Not sure if this point had been made, but the figures are for households. I think it is generally understood that why our population is still going up, household size is going down i.e, more empty nesters, fewer children, etc. This would make the figures on a per capita or per person basis even more positive and reflect even more prosperity.

Sorry to bring that news to the pessimist.

And I agree with another poster. Debt can be good or bad depending on what one does with the money. If you bought and education or a house, the expenditure will most likely eventually lead to more wealth. If one borrowed to spend in one of the new casinos or to pay one's alternative minimum tax, the result will likely be the opposite. I don't see or have the data on the expenditures, which of course did not stop the caption writer from saying the money was spent for day to day expenses.

T. Fry

Previous post:

Next post: