The Consumption Surge That Wasn’t
William Galston’s economic tour de horizon contains the following description of a recent “orgy of consumption”:
Personal consumption. Nothing has defined the past decade more than the orgy of personal consumption. From large flat-screen TVs and i-phones to furniture and foreign cars, Americans spent as though there were no tomorrow, until tomorrow came.
It’s easy to forget how unusual this period was. Once the pent-up demand from the World War Two era of austerity subsided, personal consumption was remarkably stable for three decades, averaging about 62 percent of GDP between 1951 and 1980. And then the great shift began. Personal consumption rose to 64.6 percent in the 1980s, 67.3 percent in the 1990s, and an astounding 69.8 percent between 2001 and 2008.
The thing of it is that his numbers, the ones that I quoted just there, don’t really support the characterization of a decade-long orgy of consumption at all. Here’s a decade-by-decade look at the increase in the percentage points of GDP that were dedicated to personal consumption:
The orgy of consumption, in other words, looks to have taken place in the 1980s, moderated somewhat in the 1990s, and really slowed to almost nothing in the 2000s. Nor does it seem especially reasonable to characterize the nonexistent explosion in consumption as driven by “large flat-screen TVs and i-phones.” In 2007 the average consumer unit had $63,000 in income of which $987 was spent on “Audio and visual equipment and services.” In 2000, that was $44,649 (these are nominal dollars) and $622—the share is flat.
Note that in March of 2002 you could buy a 10GB iPod for $499. Today a 120GB iPod Classic costs $249 and a 16 GB iPod Nano costs $199. Insofar as the past decade has been characterized by an explosion in cool consumer electronics, that’s because consumer electronics have been getting cheaper not because people have been drastically scaling up their expenditures on consumer electronics. Health care and education have, I believe, been driving household expenditures.
54 Responses to “The Consumption Surge That Wasn’t”
Register
RSSLogin
Topic Cloud
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
September 3rd, 2009 at 1:50 pm
But… that doesn’t support the moral narrative about how we’re all greedy pigs who deserve whatever scraps fall to us and no more!
September 3rd, 2009 at 1:51 pm
Driver’s would be health care, education, and mortgage. Home prices inflated significantly through 2007.
September 3rd, 2009 at 1:54 pm
this is a subject that would be worth digging into more, but certainly one possible explanation for the “orgy of consumption” is the increase in households as a result of demographic/family changes in the culture.
the normative ’50s family, which was a marriage-for-life concept, only had to fill one household with consumer goods.
but once those families starting divorcing and household size started falling as more people delayed marriage, we had more households to fill with consumer goods.
broadly speaking, matthew has made the key point, though: given that median household incomes have been largely stagnant throughout this period, in all likelihood, what we’re really seeing is evidence of the hedonic affect.
secondarily, btw, i suspect we’re seeing a combination of early adapters and a replacement cycle: instead of buying a new cd player when your alignment goes, buy an ipod. instead of buying a console tv when your tv is too old, buy a flat-screen.
so “ipods” and “flat screens” may well reflect an orgy of consumption without an actual orgy taking place.
September 3rd, 2009 at 1:56 pm
No, Matthew. The chart you provided is about the INCREASE in % income spent on consumer items. Rather than the 1990s “moderating” the orgy of the 80s, it actually saw an increase of 2 percentage points of income above and beyond the level spent during the “orgy” in the 1980s.
September 3rd, 2009 at 1:59 pm
Here’s a decade-by-decade look at the increase in the percentage points of GDP that were dedicated to personal consumption . . . . The orgy of consumption, in other words, looks to have taken place in the 1980s, moderated somewhat in the 1990s, and really slowed to almost nothing in the 2000s.
Why does the increase have to continue in order for the “orgy” to continue?
Let’s say the orgy starts when Matt goes from having sex with one person at a time to having sex with five people at a time. Does the orgy then stop the next hour when Matt is still only having sex with five people and not ten?
September 3rd, 2009 at 2:00 pm
By Matt’s logic, the orgy ends when he has sex with six people at a time, because the increase wasn’t as large as the last increase.
September 3rd, 2009 at 2:02 pm
This is becoming unpleasant.
September 3rd, 2009 at 2:02 pm
I have to agree with Joe and DTM here. There’s no sign that the “orgy” slowed down any from this graph.
September 3rd, 2009 at 2:03 pm
You and Galston miss the point considerably.
The issue isn’t consumption or its growth per se. The issue is the growth in household leverage to finance consumption and its growth, as a result of stagnant wages and employment in the 00’s. American household balance sheets were stressed by 2006 as the asset on which they used to leverage (their houses) began to deflate. Everyone then sought to deleverage at once, not by selling their houses, which would have reduced asset prices even further, but by reducing consumption (the paradox of thrift). That’s how we got this thing called the great recession.
September 3rd, 2009 at 2:05 pm
You seem to have completely missed Galston’s point.
Consumption growth may have slowed, but that’s very different from consumption actually going down – and Galston is talking about consumption, and how the increases of the 1980s and 1990s brought it to an extremely high level. Even if it didn’t increase that much in the 2000s, we’re still at “orgy” level, and will be until consumption actually goes down.
Slower growth is not a reduction, and the chart you’re using is a bad chart for this discussion. What you should look at is consumption year-by-year, not change in consumption decade-by-decade.
September 3rd, 2009 at 2:08 pm
I’m with Joe from Lowell. Matt, go take a statistics or policy economics course or someting.
September 3rd, 2009 at 2:09 pm
Just because the rate of increase has declined doesn’t mean Americans haven’t increased their consumerism. We definitely need national health-care and subsides to make college more affordable for the young. We also need to lower the percentage of income that is spent on consumer goods.
September 3rd, 2009 at 2:11 pm
What you should look at is consumption year-by-year, not change in consumption decade-by-decade.
Yes, but Galton also exaggerates the increase by bracketing it as he does; most of the growth between the 90s-average and the 00s-average happened during the 90s. According to that data, the percentage of consumer spending has been basically flat for the last decade.
September 3rd, 2009 at 2:16 pm
I always cringe when Matt puts a graph in his posts.
As Joe points out above, you aren’t looking at consumption, you’re looking at the increase in consumption.
If an “orgy of consumption” took place in the 1980’s, and we’ve had no decrease since, then we’re still in an orgy of consumption today.
September 3rd, 2009 at 2:16 pm
Running with the orgy metaphor, the 70s is when things began. (Cue the disco ball and music.) During the 80s the night was young and things heated up quickly. In the 90s the tempo continued to increase, but the rush was slower than in the 80s. The 2000s started then at an orgy almost ready to climax, but all resources were needed to keep it going until everyone collapsed, and the orgy was over.
September 3rd, 2009 at 2:21 pm
Still, I’m fairly sympathetic to the thesis that said level of consumption is extreme. After all, we are significantly off historical norms in terms of consumption, not to mention debt and savings. So while he may have exaggerated the data, I think even the unexaggerated data supports his argument.
September 3rd, 2009 at 2:22 pm
Health care and education have, I believe, been driving household expenditures.
As well as the cost of the house itself.
September 3rd, 2009 at 2:24 pm
When anything reaches 69 percent, it’s an orgy.
September 3rd, 2009 at 2:36 pm
Matt, as several above have said, you’ve totally blown this one. Time for another post agreeing and apologizing. (Or are you a Villager now?)
September 3rd, 2009 at 2:44 pm
I am unsure whether this chart proves the point- considering health ate up a lot of GDP….. blah blah blah… I see the point is stated- took too long to post.
September 3rd, 2009 at 2:52 pm
Look, folks. If the economy is getting better now because we’re losing jobs at a slower rate than we were losing jobs in the spring, then the orgy is moderating because consumption is increasing at a slower rate now than in the prior years.
September 3rd, 2009 at 2:53 pm
It’s an orgy of healthcare. That can’t be good for you, can it?
September 3rd, 2009 at 2:58 pm
Ditto what others have said about consumption level v increase in those levels.
I think that if the graph argues for anything its that the level of consumption we’ve had in the last ten to fifteen years, over 75% of the GDP, is probably the maximum any economy could possibly have. Hence the flat level since the 90s.
September 3rd, 2009 at 2:58 pm
From what I see, Matt is basically right. Consumption hasn’t been that much higher during the 00s than in the 80s or 90s. You can say that consumption levels of 65% of GDP and higher are too high, but that’s not the same as “Nothing has defined the past decade more than the orgy of personal consumption”.
If you change that to “Nothing has defined the last 30 years more than the orgy of personal consumption”, but I guess it doesn’t has the same alarmist ring to it.
September 3rd, 2009 at 2:59 pm
Ack, should read “65%”
September 3rd, 2009 at 3:00 pm
then the orgy is moderating because consumption is increasing at a slower rate now than in the prior years
No Al, you’re making the same mistake Matt did.
If a person eats 5000 calories a day, then he’s gorging himself.
If he’s still eating 5000 calories a day a year later, he hasn’t “moderated his consumption”, he’s still gorging himself.
September 3rd, 2009 at 3:02 pm
The orgy of consumption, in other words, looks to have taken place in the 1980s, moderated somewhat in the 1990s, and really slowed to almost nothing in the 2000s.
In other words consumption had reached unsustainable levels by the new millennium, and has continued to inch upwards from there over the last decade.
September 3rd, 2009 at 3:22 pm
Elizabeth Warren has been all over this, in detail and definitively. Since the 80s the price of consumer goods have drastically abd the price of necessities (housing, medical, college) has skyrocketed.
There was a hour long video with charts posted at Mark Thoma’s.
September 3rd, 2009 at 3:27 pm
I think bob mcmanus was missing a word, but I think his point was that the cause of this increase in consumption wasn’t increased spending on low priority items, it was increased spending on things like housing, health care, and education. Which is my understanding as well.
September 3rd, 2009 at 3:28 pm
Coming Collapse of the Middle Class …Youtube, 57 minutes
Maybe there is a transcript somewhere.
The point is consumption may have declined due to cheap imports without any real improvement in middleclass security.
September 3rd, 2009 at 4:11 pm
After all, we are significantly off historical norms in terms of consumption,
Except that this way he’s portraying the 50s, 60s and 70s as ‘normaal’ and then labelling an equal span of time as aberrant. It would be as fair to say that the 50s, 60s and 70s were the aberration– which, of course, they were, because real median incomes were actually increasing. Thus, consumption could increase in real terms but not increase as a percentage of income.
September 3rd, 2009 at 4:45 pm
Matt’s post is very confused. A slower rate of increase in consumption is not a moderation of consumption. It’s only a moderation of the growth of consumption. His claim about consumer electronics is similarly confused. The dramatic decline in the price of consumer electronics allowed people to consume much more consumer electronics without increasing the share of their income they spend on consumer electronics.
September 3rd, 2009 at 4:52 pm
I’m afraid MY’s critics are right. His chart shows first differences, not levels of activity. It suggests that the orgy didn’t continue to become more & more frenzied, not that it ended. That said, it is worth focusing on what changed in the 1980s. Popular conservative narratives of the period were always based on a misconception, but until now much of the public has been unreceptive to better-grounded counternarratives. Maybe recent events will change that.
September 3rd, 2009 at 4:52 pm
Since the 80s the price of consumer goods have drastically abd the price of necessities (housing, medical, college) has skyrocketed.
Utter nonsense. If the (real) prices of those necessities have “skyrocketed,” but real incomes have not, how is it that people today consume far more of them than people did in the 80s? Houses are bigger. Health care is better. The rate of college enrollment is higher than ever.
September 3rd, 2009 at 4:59 pm
Why do Matt’s graphs always suck? It’s almost as if he’s an innumerate fool.
September 3rd, 2009 at 5:07 pm
If the (real) prices of those necessities have “skyrocketed,” but real incomes have not, how is it that people today consume far more of them than people did in the 80s?
More borrowing and less saving. A trend that was bound to be unsustainable–and it was.
Of course it isn’t literally “people today”–people today have cut consumption, increased savings, and are paying down debt.
September 3rd, 2009 at 5:26 pm
More borrowing and less saving. A trend that was bound to be unsustainable–and it was.
The long-term trend in household net worth has also been positive. Try again.
September 3rd, 2009 at 5:33 pm
The long-term trend in household net worth has also been positive.
And that also just proved unsustainable: we’ve crashed back to roughly the same level of household net worth as a percentage of GDP we had back in the early 1960s. So we were borrowing against assets that weren’t actually worth what we thought they were worth.
September 3rd, 2009 at 5:44 pm
And that also just proved unsustainable: we’ve crashed back to roughly the same level of household net worth as a percentage of GDP we had back in the early 1960s.
You are confused. Real GDP per household has grown enormously since the early 1960s. So even at a constant share of GDP, real household net worth has also grown enormously. People today both consume more housing, health care and college education than people did in the early 1960s, and they have a greater real net worth than people did in the early 1960s.
September 3rd, 2009 at 5:55 pm
I think we can all agree that the big increase in consumption came in the 80’s (Reagan) grew somewhat in the 90’s (Clinton), and flattened in the 00’s (Bush). Whether that level is an historic aberration or objectively high (or deserves the loaded appellation “orgy”) cannot be discerned from this graph or any information in either column.
But whatever it was, Reagan started it.
September 3rd, 2009 at 8:09 pm
Real GDP per household has grown enormously since the early 1960s. So even at a constant share of GDP, real household net worth has also grown enormously.
We’re talking about personal consumption as a percentage of GDP and how it could increase without an equivalent increase in personal disposable incomes. The answer is that we financed it out of more borrowing and less saving.
You then responded we had rising household wealth and said I needed to “try again”. That rise in household wealth turned out to be a mirage as a percentage of GDP, which is what is relevant for the sake of this discussion because we are talking about increases in personal consumption as a percentage of GDP.
So again, we funded increased personal consumption as a percentage of GDP out of decreased savings and increased borrowing. We thought we would be OK doing this because on paper we had increased household wealth as a percentage of GDP. But that increase of household wealth as a percentage of GDP turned out to be a mirage.
September 3rd, 2009 at 8:20 pm
Matt updates:
I agree blaming spending on gadgets is silly, and I agree that the “orgy of consumption”–defined as an unsustainable consumption rate in light of the share of GDP going to personal disposal income–didn’t just start a decade ago and indeed didn’t get much worse over the last decade.
But what I was objecting to is the idea that this orgy of consumption “moderated somewhat in the 1990s, and really slowed to almost nothing in the 2000s.” Rather, the orgy was still going on, as demonstrated by decreasing personal savings rates and increasing household debt through this period. And it took the Great Recession to finally bring the orgy to an end.
September 3rd, 2009 at 8:41 pm
We’re talking about personal consumption as a percentage of GDP
No, we’re talking about Bob McManus’s claim that the price of housing, health care and college education has “skyrocketed” since the 80s. I asked how it could be that the consumption of those things has increased so much if income growth has not kept pace with those allegedly “skyrocketing” prices. You responded that the growth in consumption has been financed from savings and debt. Your claim is contradicted by the fact that household net worth has also grown since the 80s. This has nothing to do with household net worth as a percentage of GDP. It’s about real prices, real incomes and consumption. Try again.
September 3rd, 2009 at 8:48 pm
Careful with how housing is treated both in the GDP figures (all housing consumption is imputed rental costs) and in the BLS CES Matthew is quoting here (only cash flows count — option ARMs are low cost housing according to CES). Lot of costly housing investment that people are now stuck with doesn’t show up as housing consumption or expenditure.
Getting these numbers right would be interesting.
September 3rd, 2009 at 9:40 pm
You responded that the growth in consumption has been financed from savings and debt. Your claim is contradicted by the fact that household net worth has also grown since the 80s.
I just linked graphs showing that as consumption as a percentage of GDP increased, the personal savings rate went down, and household debt as a percentage of GDP went up. Meanwhile, I also linked a graph showing that most of the supposed increase in household net worth as a percentage of GDP turned out to be a mirage.
Whatever economic model you have in your head is contradicted by the actual facts.
September 3rd, 2009 at 9:49 pm
By the way, as far as I can tell Mixner’s basic conceptual error is that he doesn’t realize that household net worth is a function not just of liabilities but also of assets, and that valuing assets at what proved to be inflated prices is what made it possible for household net worth and household debt to be increasing at the same time. Again, though, those inflated asset prices proved unsustainable, and thus so did those valuations, and thus so did the prior calculations of household net worth.
Which leaves dealing with all that household debt that wasn’t backed with as much in assets as we thought. So now some of that is debt is being dealt with through foreclosures, bankruptcy, work outs, and so on, and the rest is being dealt with through an increase in the savings rate. All of which isn’t doing the economy any good, but it is a necessary step toward finally getting back on a sustainable economic path.
September 4th, 2009 at 12:24 am
I just linked graphs showing that as consumption as a percentage of GDP increased…
You’re still confused. Consumption as a percentage of GDP is irrelevant. You claimed the growth in consumption was funded by savings and debt. The fact that net worth increased contradicts your claim. Try again.
September 4th, 2009 at 12:39 am
By the way, as far as I can tell Mixner’s basic conceptual error is that he doesn’t realize that household net worth is a function not just of liabilities but also of assets, and that valuing assets at what proved to be inflated prices is what made it possible for household net worth and household debt to be increasing at the same time.
More nonsense. Real household net worth has been increasing for decades. It’s not a matter of inflated asset prices from short-term housing or stock market bubbles. It’s a long-term trend of growing real household wealth over generations. And it completely contradicts your claim that the growth in consumption of housing, health care and college education has been funded by drawing down savings and taking on uncollaterized debt. If that had been the case, real household net worth would be lower now than it was in the 80s, not higher. You don’t know what you’re talking about. As always.
September 4th, 2009 at 2:20 am
Consumption as a percentage of GDP is irrelevant.
It is the subject of this thread.
You claimed the growth in consumption was funded by savings and debt.
It was. See the charts I linked.
It’s not a matter of inflated asset prices from short-term housing or stock market bubbles.
The effects of the dot-com bubble and housing bubble on household net worth were both readily apparent. See the chart I linked.
September 4th, 2009 at 2:55 am
It is the subject of this thread
It is not the subject of the question of mine you attempted to answer. It is irrelevant to that question. Your answer is wrong for the reasons I have explained.
It was. See the charts I linked.
No, it wasn’t. The charts you linked to do not support your claim that the growth in consumption was funded by savings and debt. The fact that household net worth has increased disproves your claim.
The effects of the dot-com bubble and housing bubble on household net worth were both readily apparent.
The dot-com and housing bubbles were short-term events. The increase in household net worth is a long-term trend that disproves your claim that the long-term increase in consumption of housing, health care and college education has been funded by savings and debt. If your claim were true, the long-term trend in household net worth would be negative. It isn’t negative. It’s positive. Your claim is false. Try again.
September 4th, 2009 at 7:39 am
It is not the subject of the question of mine you attempted to answer.
Your question was about bob’s statement. bob’s statement was about the topic of the thread.
The fact that household net worth has increased disproves your claim.
It hasn’t increased versus the early 1960s, and even if it had it wouldn’t disprove the claim because net worth is a function of both assets and liabilities. And we know household debts as a percentage of GDP increased. So we don’t need try to to deduce household debt levels from household net worth.
The dot-com and housing bubbles were short-term events.
When their effects on asset prices disappeared it turns out that household wealth as a percentage of GDP was at that same level as the early 1960s.
September 4th, 2009 at 11:30 pm
Now when it comes to the middle class overspender, suddenly the tone is nice and defending. But imagine a teacher Union might try to ensure some job safety. All up in arms! Yes, rising income inequality and healthcare/college tutotion inflation didnt help, yes bad regulation didnt help either. But at the end of the day, not only the Bush government, but also many if not most Americans were reckless spenders.
September 5th, 2009 at 8:26 am
[...] 65% czegoś co statystycznie jest konsumpcją indywidualną. Potwierdzają to konkretne dane, na przykład o wydatkach na sprzęt audio-video. W 2000 roku przeciętne gospodarstwo w USA miało $44 649 dochodów i wydało na sprzęt RTV $622 [...]
September 8th, 2009 at 1:35 pm
I’m guessing there’s a selection bias problem with the chart above, in that growth during the 1980s was high at least in part because we entered the 1980s at/near the low point of the business cycle, and that growth during the 00’s is low because we entered the decade at/near the high point. A more rigorous data analysis is required for any conclusions to be reached.