The US economy shrank at an annualized rate of 2.9 percent in Q1

http://www.washingtonpost.com/blogs/wonkblog/wp/2014/06/25/economy-shrank-almost-3-percent-in-q1-holy-guacamole/

by Bahamut • 12 years ago

79 93 12 years ago

washingtonpost.com

sp3323• 12 years ago

The economy shrank at an annualized rate of 2.9% over one quarter. That's 2.9% / 4 ~= 0.7% in Q1.

smrtinsert• 12 years ago

Could you explain this please?

jerf2• 12 years ago

"If every quarter this year was the same as this one, it would be a 2.9% decline for the year."

sp332 uses the ~= symbols for approximately equal because technically you can't just divide the percentage by four like that. 2.9%/4 = .725% (exactly), but technically four quarters at .725% shrink each would produce 100%-((100%-.725%)^4) for the year, which is a 2.869...% decrease, not exactly 2.9%. But for small percents being compounded a small number of times it is a decent approximation, in this case thoroughly dominated by measurement inaccuracies and noise levels; the small percentage and the small number of compoundings work together to keep the error very small.

kinghajj• 12 years ago

Meaning that if the rate of decline experienced in Q1 continued throughout the year, then the annual shrinkage would be about 3%. 1.0074^4 is about 1.03, so the actual shrinkage for the quarter in question was 0.74%.

logicallee• 12 years ago

if the economy shrinks by 0.05% on the Sunday before thanksgiving, you could annualize that by multiplying by 365 and getting an 18.25% shrink in the size of the American economy on that day.

It would be an extremely false statement to write, "The U.S. Economy contracted by 18.25% on Sunday before Thanksgiving", which is how our title reads. ("The economy shrank almost 3 percent in Q1").

The number 3% doesn't appear anywhere, just as the number 18.25% doesn't appear anywhere. sp332's very good point is that it should include the word "annualized" or read "0.7%". A huge difference!

badman_ting• 12 years ago

I remember idle moments during soccer games I played as a kid. I would sometimes think to myself "OK, we're up 1-0. If the rest of the game goes at this rate, we will definitely win." The problem with that reasoning is the word "if".

clintonc• 12 years ago

Math pedant to the rescue! Actually, rates combine in a multiplicative way -- losing 3% and then 4% does not lose 7%, (though it's close). What if the economy lost 1% each day for a year? It would not lose 365%.

If we assume the annualized rate, the economy would be 97.1% of its current size in a year. The fourth root is 99.267%, so the economy shrank about 0.733% this quarter. If it does so for four consecutive quarters, it will decrease by 2.9%.

I know that's what you said in the first place, but when extrapolating further into the future, it makes a difference.

sp3323• 12 years ago

Actually I didn't think about annualized rates compounding like that, although it makes sense now that you mention it :)

conistonwater• 12 years ago

Math pedants of the world:

If you calculate 1-(1-x)^(1/4), the difference between the true answer and x/4 is approximately 3 * x^2/32. For a figure of y percent, this is a relative error of 0.375 * y percent. Ignore it when y is small.

The GDP figure is only known to something like 1 or 2 decimal places.

deadfish• 12 years ago

Whose economy?

wutbrodo• 12 years ago

The article is from the Washington Post so you can probably assume it's the US economy.

dang• 12 years ago

Since the comments so far have been only about the title, we edited the title to address these objections.

memossy• 12 years ago

The most interesting component of this was the healthcare drag in the revisions taking it down from the initial ones.

While a large element of this was the affordable care act, it is also notable that healthcare inflation went negative for the first time in 20 odd years last quarter (education did too, but thats another story).

Given the US currently spends twice the amount of other countries on its health care as a % of GDP (14%) for similar health outcomes and individuals are being significantly disincented to spend on healthcare via subsidies and high deductibles respectively, I think we will see increasing normalisation of healthcare spend to levels under 10% of GDP, a huge drag on GDP.

This will be further accelerated by smartphone and biometric proliferation allowing for easier and more reliable diagnosis and treatment choice.

Interesting times.

refurb2• 12 years ago

Unfortunately the ACA doesn't do much to address increasing healthcare costs. There are a few initiatives (comparative medicine funding is one that comes to mind) that are good starts, but the ACA is likely to lead to overall higher GDP expenditures on healthcare in the US, not lower, as more and more uninsured have access to care.

clavalle2• 12 years ago

Your analysis assumes healthcare is like any other commodity.

While the primary aim of the ACA is not overall cost reduction (but, rather, make it affordable for the uninsured to get insurance), there are several reasons it could go down that path.

The marketplaces and minimum standards for insurance plans make it easier to comparison shop.

The fact that people will have insurance will lead to earlier and cheaper treatments for potentially serious disease.

Businesses can get out of the insurance business altogether. As businesses remove themselves from the equation individuals will be more attuned to the cost of their insurance plans and shop accordingly.

There are some direct provisions for paying providers for disease prevention and effective treatments that lead to patients not having to return for the same issue rather than just for treatments reversing that twisted financial incentive.

Granted, there are plenty of ways for these potential cost benefits to go sour but there is a good chance that the ACA will actually reduce costs relative to what they would have been without the ACA.

rayiner3• 12 years ago

The healthcare example is a very good illustration of why the GDP calculation is susceptible to the broken window fallacy: http://en.wikipedia.org/wiki/Parable_of_the_broken_window.

The root of the problem is that GDP doesn't measure changes in capital stock. For example, after Fukushima, Japan's GDP ticked up slightly because of the expenditures from disaster recovery. Obviously it's absurd to conclude that Fukushima was good for the economy--the loss of capital more than outweighed the money spent fixing things. The same principle has many different implications. For example, if we reduced air pollution, expenditures on treating lung cancer would go down, hurting GDP. Countries that are heavily based on oil or mineral exports have artificially inflated GDP, because the measure excludes the decline in the value of reserves.

bengali32• 12 years ago

Interesting point regarding the parable, thank you for sharing.

If GDP does included anything stock, would that really be accurate? Not an expert, but isn't stock priced on the future value of something (ie dependent upon future sales forecasts). since GDP is backward looking, i don't think it should be counted? stock could be used as a predictor of future GDP since the stock is trying to put a value to future sales.

Also, wouldnt that also double count certain sales? ie each iPhone sold would increase goods sold as well as the companies stock price?

Zigurd2• 12 years ago

I think most people here would agree with that.

But getting a true measure of the ability of the US economy to create a comfortable life for as many people as possible also needs to take a very hard look at military spending and the share of the manufacturing economy that goes into actual and potential destruction of capital. Are those police rifles and APCs "economic output?"

When the Soviet Empire was stripped of an outsize military and internal security apparatus, what was left was a GDP the size of Italy.

So it cuts both ways: Health care deflation is probably good, and calls for a higher quality of life number. But that monstrous security state we are dragging around also should get moved to the negative column.

mtdewcmu2• 12 years ago

Obviously it's absurd to conclude that Fukushima was good for the economy--the loss of capital more than outweighed the money spent fixing things.

That depends on your definition of "economy," doesn't it? When I picture an economy, I picture goods and services being consumed and money changing hands. That's what GDP measures. Should money locked up as capital be counted as part of the economy?

SilasX• 12 years ago

Normally, I make the same point[1], but I think you're falling into the opposite mistake here: if you're using GDP as a proxy for economic activity, regardless of cause, then it is correctly capturing the insight that "people started doing more stuff as a result of the disaster [though they may be poorer]". This (usefully) distinguishes it from e.g. "people were impeded from rebuilding, and so continued on as before".

The problem you allude to, rather, is that:

1) People use GDP as a proxy for some kind of "economic goodness" (not simply "activity").

2) The two concepts generally correlate.

3) But there are known cases where the two diverge -- when economic badness happens, and yet GDP goes up.

4) But people still call it good despite the recognition of the special case.

[1] I think the appropriate way to economically characterize window-breaking/repair scenarios is: "It used to be hard to identify good uses of scarce resources; now that vital stuff got destroyed, it's easy." See: http://blog.tyrannyofthemouse.com/2011/12/broken-windows-par...

merpnderp• 12 years ago

Similar outcomes? Maybe you should compare cancer and heart disease survival rates (the two biggest killers in the US).

The US does an incredible job with a fat, unhealthy, diverse population, even if we pay dearly for it.

mikeash3• 12 years ago

Can anybody who understands this stuff better than I do comment on the winter weather factor?

To me, the "winter sucked" excuse sounds like complete BS, a transparent lie along the lines of, "Well, I just didn't want to order the tide to stop right now." This is a massive change (down a total of 7% or so from two quarters prior) and surely the weather, while unusual, wasn't that strong.

But maybe it really is a big factor. I'm far from an expert. Can anyone comment on whether the weather (heh) reasoning is sane?

pyre• 12 years ago

Well, the effects of the winter can be:

- Damage due to harsh weather causing a net drag on businesses, causing them to be more conservative with spending.

- All of the store, school, etc closures due to inclement weather.

- Even without store closures, many people are less likely to leave their homes (i.e. spend money) when the weather is bad, even if the weather isn't 'snow day' bad.

I feel that it's probably unlikely to be entirely due to the weather, but I don't think that we can discount the effects a harsh winter can have.

btilly• 12 years ago

Let's run some fake numbers to see.

A quarter is roughly 91 days. About 65 of which are week days, and that is where most economic activity happens.

Suppose that bad weather costs you 30% of economic activity in half the country for just one of those week days. That works out to losing 0.23% of the quarter's economic activity. But we are using a quarter to estimate annual growth, so that now looks like losing 0.92% annualized growth. From one day of disruption over part of the country.

As you see, blip from a bad winter storm on the East Coast really can cause the economy look like it is headed in a much worse direction than it is for a quarter.

gone35• 12 years ago

It's not intuitive, but since building sites must halt work during strong winter days, a large increase in the number of 'snow days' has an outsize effect on the total output of the construction sector --one of the largest sectors of the US economy-- during Q1.

This past winter was peculiar in that regard because not only there were more snowstorms overall than the previous year, but also there were major snowstorms in regions that seldom see that kind of weather like the South. And it shows in the figures for residential and commercial building during the quarter.

These seasonalities however are well understood and don't matter nearly as much, in part because construction companies often catch up on their delayed projects during the subsequent quarter. This is one the reasons why reading too much into annualized quarterly fluctuations can be misleading at times [1].

[1] http://jaredbernsteinblog.com/whoa-whassup-with-that-big-neg...

mikeash3• 12 years ago

Interesting point about catching up on work in the subsequent quarter. This sort of thing is why I thought the impact of weather was overstated, but I thought that the catchup would have happened during the same quarter. If it takes just a little bit longer then that would skew things a lot.

Also, I love how I can ask this question and get a ton of quality answers in quick succession.

rz2k• 12 years ago

There are a few adjustments to the data for getting a better impression of the data. For example "real" data attempt to minimize the distortions that come from changes in prices, while the raw "nominal" data measure economic activity. Another adjustment is for the seasonality of data. Here's a graph of some a few retail data series, where not seasonally adjusted appear with the seasonally adjusted data. [1]

Here you can take a look at the most recent release of the Employment Situation.[2] It states that the survey of businesses indicated 217,000 more people working. Now, look at the table of the data.[3] You have

May: 139,192 thous Apr: 138,272 thous

=> Surveys of businesses therefore indicated that 920 thousand more people were working!

So, what's going on? It is that no one cares about reporting the seasonal cycles in employment, because it isn't news that helps you understand economic trends. So, people actually look at the seasonally adjusted data.

May: 138,463 million Apr: 138,246 million

=> 217 thousand more people

In order to come with the seasonal factors the Census (and BLS and BEA) use an autoregressive integrated moving average. Back to your original question, what does it mean to say that winter sucked? It means that people's prosperity was lowered by the winter, just like it always is, but while the seasonal factors usually remove that effect this time they are part of the seasonally adjusted data because the seasonal factors are derived from past data with less severe winters.

Really it comes down to the fact that a single score is inappropriate for an understanding of the economy. GDP measures GDP, and something else with a different methodology would measure something else. You can find the fallacy acted out online with sites that purport to improve the data with adjustments without explaining the tradeoffs that come with adjustments by definition.

[1] http://research.stlouisfed.org/fred2/graph/?id=RSCCAS,RSCCAS...,

[2] http://www.bls.gov/news.release/empsit.nr0.htm

[3] http://www.bls.gov/news.release/empsit.t17.htm

maerF0x0• 12 years ago

Seems to me the idea of declaring a shrinking economy in terms of $ doesnt make sense. What if the average price of things is dropping (eg Healthcare)?

To me it seems the goal should just be how much stuff did we make? If that number went up, then good (except corn).

jiggy2011• 12 years ago

How would you define "stuff" and "make"? Especially with regards to financial services etc.

imaginenore• 12 years ago

When it comes to the financial services, we can simply measure the cocaine consumption.

pessimizer11• 12 years ago

Seems to me the idea of declaring a shrinking economy in terms of $ doesnt make sense. What if the average price of things is dropping (eg Healthcare)?

We know what inflation is, and that the price of healthcare is not dropping.

To me it seems the goal should just be how much stuff did we make? If that number went up, then good (except corn).

We also know about the trade deficit: http://www.businessinsider.com/us-trade-balance-march-2014-2...

bengali32• 12 years ago

Do you mean the total creation of goods in the US? To compare apples to oranges (literally) we compare their dollar values and call it GDP. comparing the 'count' of things wouldn't be as useful as we could make a whole bunch of things that no one wants and call it productivity.

lotsofmangos• 12 years ago

Why do you think a rise in corn production is bad?

Also, in what units are you going to measure the total amount of stuff made, if not in currency?

pessimizer11• 12 years ago

A stock market recovery is not a "recovery."

Prime-age employment: http://data.bls.gov/timeseries/LNS12300060

Median wealth, and the median wealth of various segments: http://finance.yahoo.com/blogs/daily-ticker/for-most-familie...

Median income: http://advisorperspectives.com/dshort/updates/Median-Househo...

Wealthy people cannot drive consumption, because they spend a far smaller part of their income on it.

edit: I intentionally use the prime age employment rate (as is convention of late) in order to avoid objections that the low employment rate is the fault of a large segment of baby boomers retiring.

The general numbers look worse, not better: http://data.bls.gov/timeseries/LNS12300000

sillysaurus32• 12 years ago

That unemployment graph is rather breathtaking, both because it illustrates how impactful the '08 crash was, and how little we've recovered since then. Is there any chance that the graph could be misleading, or should it be believed at face value that one in four people are unemployed?

opendais2• 12 years ago

The graph represents a 4% chance in the size of the active labor force.

In good times, 1 in 5 people don't work. So saying 1 in 4 is a massive change [when it is 5%] isn't really correct.

It also is only 'prime' so 25-54. 55-65, people still work. Same with under 25.

michaelbuckbee• 12 years ago

One potentially misleading aspect of the chart is that it covers only 25-54 year olds and one result of the downturn was that older workers have remained in their jobs longer, delayed retirement, etc.

dllthomas• 12 years ago

I think this includes people who are not working by choice (wealthy and idle, stay-at-home parent, student, &c).

nostrademons4• 12 years ago

The graph doesn't start at 0 on the Y-axis. It would be a lot less striking if the scale ran from 0-100 instead of the cherrypicked 74-81 scale.

Also, this is labor force participation, not unemployment. If you carried this back to the 1950s, it would be around 60%, because most women didn't work back then.

http://research.stlouisfed.org/fred2/series/CIVPART

endersshadow• 12 years ago

This is an employment to population ratio, so 1 in 4 people, ages 25-54 do not work. That does not mean that they are unemployed! Unemployed people are those looking for work that don't have employment. Plenty of people voluntarily leave the "labor force" (employed + unemployed people), such as stay-at-home parents, trust fund kids, or independently wealthy and retired folks.

That said, "unemployment" is a very tough figure to calculate (as are most macroeconomic figures), and there's a lot of pseudo-controversy about what the numbers mean. Wiki does a good job summarizing how the BLS calculates unemployment [1].

[1] https://en.wikipedia.org/wiki/Unemployment#United_States_Bur...

DangerousPie2• 12 years ago

Very misleading Y axis. Here is the same plot with a proper axis: http://i.imgur.com/k4ClDSF.png

cs7022• 12 years ago

TL;DR: annual US GDP declined 2.9% because consumers spent less. In fact, they are not yet spending at rates anywhere near the rates at which they were spending before the financial crisis in 2008.

When consumers spend less, businesses in the aggregate sell less of everything, because every dollar spent by a consumer is a dollar earned by someone else -- usually a business.

Before the financial crisis, consumers borrowed aggressively to finance consumption, like drunken sailors... but unlike the financial sector, they never got a bailout. They were forced by the circumstances to follow Steve Martin's advice from Saturday Night Live: "don't buy stuff you cannot afford."[2]

Are we really that surprised that many consumers are reluctant to borrow and/or spend like before the crisis?

Edits: modified first paragraph to convey what I actually meant to say, in response to ctl's comment. The original paragraph was poorly written. (Thanks for pointing out the inconsistency, ctl!)

[1] http://www.npr.org/blogs/money/2009/02/household_debt_vs_gdp...

[2] https://screen.yahoo.com/dont-buy-stuff-000000884.html

executor142• 12 years ago

Actual TLDR:

Spending was muted because of the harsh winter. Recent economic indicators have been very positive so expectations are for a significant rebound for Q2.

itbeho• 12 years ago

Looks to me like it's how the government is accounting for spending in healthcare that is the biggest contributor to these numbers. Next quarter they will report big gains, just in time for campaign season and the run up to the election.

http://www.businessinsider.com/why-health-care-spending-in-g...

crazy1van• 12 years ago

Are we really that surprised that many consumers are reluctant to borrow and/or spend like before the crisis?

Seems like it is not only unsurprising but also a good thing. People learned a lesson about spending recklessly and some economists would claim that's a bad thing. GDP is not the end-all, be-all metric of a healthy economy.

adsenseclient• 12 years ago

The main thriving product of the 21st century American economy is Stock of publicly traded US companies. The stock market is largely decoupled from economic reality. As long as the total market cap of all US companies is going up (which was the case in Q1), there is a "recovery". The settlement between this recovery and the real economic decline will only occur around 2040, when US dollar is debased and finally ceases to serve as the world reserve currency. Feel free to downvote.

dclowd99012• 12 years ago

Part of this slowdown in consumer spending also has to do with the tightening of credit as well, making it more difficult for people without means to spend. In my opinion this is a great thing. When people are backed against a wall because they have no other way to pay for their livelihood than their salary, they're not going to be ok with stagnant wages and companies low balling them.

Bring on the pain, I say. Real growth comes from companies investing in their workforce, not people drowning themselves in debt.

Put another way: it used to be that you could buy a house for the equivalent of a year's salary. Now it seems most houses are about 5-6x the average salary of a given area. If that were still the case today, most Bay Area residents should be making near 1 mil/year or houses here should be closer to about $150-200k, more likely the latter since credit is primarily the reason for the inflated cost.

crassus• 12 years ago

I know that people will make political hay of this, but there is danger to managing to quarterly statistics. If your economy needs some adjustments to achieve long-run sustainability then it's possible that GDP may fall for a quarter or two and there's nothing wrong with that.

GDP is a great metric for long-term or inter-national comparison. It's terrible as a quarterly national management metric.

malandrew4• 12 years ago

Interesting observation. It's really hard to shake off the shackles of presentism and imagine what the US would look like on an alternative timeline where the US government had not aggressively promoted home ownership and easy credit. The economy and legislation are now so thoroughly distorted in favor of home ownership that it's difficult to imagine how things could have been different.

Specifically, I'm wondering if there was a time when we were having the same debate about housing that we are now having about college tuition, but that it has been long enough (and there have been economic gains for homeowners) that we conveniently forget how much our longstanding policies drive up housing prices. The idea of buying a home on a full year's salary sounds crazy, but then again there was a time when 4 years of college could be paid off fully via part time work while at school, something that is distant pipe dream these days for the majority of jobs within the reach of most college students.

zghst• 12 years ago

I wouldn't say that credit is driving up the costs of Bay Area homes, but the amount of wealthy workers in the area.

And also consider that homes built today have to go through tons of regulatory costs (building codes, politics with permits), there are new technologies installed, AC, internet, upgraded roofing, better foundations, upgraded wiring, possibly an HOA, rec center, etc, paved sidewalks in front of every house. The US Census Bureau has also determined that the average size of a home is 2,480 square feet vs 1,600sqft in 1970. All the progress that we have made with homes today costs.

pessimizer11• 12 years ago

The funny thing is that consumption is actually historically very, very high:

http://www.tradingeconomics.com/united-states/personal-savin...

Savings rate, now: 4%

Pre stock-bubble average: 8%

edit: "Personal Savings in the United States increased to 4 percent in April of 2014 from 3.60 percent in March of 2014. Personal Savings in the United States averaged 6.82 Percent from 1959 until 2014, reaching an all time high of 14.60 Percent in May of 1975 and a record low of 0.80 Percent in April of 2005."

jonnathanson2• 12 years ago

A lower Personal Savings Rate doesn't necessarily mean people are "consuming" more. At least not the sort of consumption (discretionary) that many businesses care about. It simply means that the proportion of income being saved is lower. There are any number of factors that could make that true, including rising costs of living or inflation vis-a-vis stagnant income.

In fact, while personal income has been increasing on a low nominal basis, income growth rate has been slowing.

Yes, PSR accounts for "disposable" income, but that term can be somewhat misleading. Disposable income is simply the net of income minus taxes. It's not necessarily available for discretionary purposes.

I would say that savings rates are quite low right now, and that accordingly, nominal consumption is high. But there's a deeper story there.

http://www.bea.gov/newsreleases/national/pi/pinewsrelease.ht...

jcampbell12• 12 years ago

The decline is seasonally adjusted vs Q4 2013. Since both periods are after the recession, your TL;DR make no sense, and is a misleading interpretation.

ctl• 12 years ago

Your comment makes no sense. GDP is slightly down compared to 2013. Why are you talking about 2008?

The article implies that the main driver here is reduced consumer spending from a harsh winter.

cs7022• 12 years ago

Why are you talking about 2008?

I'm talking only about consumer borrowing and spending prior to the crisis in 2008. My understanding is that, five years after the crisis, consumer spending still has not recovered[1], which seems rather unusual for an economic recovery.

PS. I modified the first paragraph in my comment above, so it more clearly conveys what I actually meant to say.

[1] For example, see http://www.clevelandfed.org/research/trends/2014/0214/01houc...

snake_plissken• 12 years ago

Real interest rates are negative and credit is tight. Credit has been tight for some time, but there is some debate as to when the real interest rate went below zero. These two factors will cause the US economy to (eventually) shrink because ~65% of its component parts reflect economic activities based on consumption. The economy can not expand because dollars that have historically gone into consumption are being invested into Treasuries. This has the effect of keeping yields on treasury notes down, which causes real interest rates to go below zero. A negative real interest rate environment implies investors are willing to take a small loss if it means they can avoid bigger losses on other investments.

muzz• 12 years ago

What does this article have to do with Hacker News?

Note the decidedly different tone from this entry on the same Washington Post Wonkblog:

http://www.washingtonpost.com/blogs/wonkblog/wp/2014/06/25/t...

esbranson2• 12 years ago

Everything and nothing. Nothing to do with computers, everything to do with hype and name recognition.

Besides, computers are just tools. A means to an end. To the hammer designer, is a discussion about home building relevant? The economics of home building? The economics of people buying homes? The psychology of homebuyers? Is psychology relevant to hammer design? Maybe. Relevance is complicated.

joesmo• 12 years ago

"Recession-like. It is astounding that we are using any variation of that word five years into the recovery."

No, it's not. There is little to no recovery. The only surprising thing about analysis like this is how far the author goes to avoid explicitly stating this when it's so implicit in the article.

The US economy shrank at an annualized rate of 2.9 percent in Q1

Related Stories