Gross Domestic Error

Pierre Lemieux at EconLib catches a surprising error from The Economist which wrote this week:

There was some head-scratching this week, as data showed Japan’s economy growing by 2.1% in the first quarter at an annualised rate, defying expectations of a slight contraction. Most of the growth was explained by a huge drop in imports. Because they fell at a faster rate than exports, gdp rose.

Nope. Imports do not influence Gross Domestic Product, at least not in the mechanical way suggested by The Economist. Here’s how Tyler and I explain it in Modern Principles:

It’s important to remember the Domestic in Gross Domestic Product. When we add C+I+G we are adding up all national spending but some of that spending was on imports, goods that were not produced domestically. So we subtract imports from national spending to get national spending on domestically produced goods, C+I+G – Imports.

…Here is a mistake to avoid. The national spending approach to calculating GDP requires a step where we subtract imports but that doesn’t mean that imports are bad for GDP! Let’s consider a simple economy where I, G, and Exports are all zero and C=$100 billion. Our only imports come from a container ship that once a year delivers $10 billion worth of iPhones. Thus when we calculate GDP we add up national spending and subtract $10 billion for the imports, $100-$10=$90 billion. But suppose that this year the container ship sinks before it reaches New York. So this year when we calculate GDP there are no imports to subtract. But GDP doesn’t change! Why not? Remember that part of the $100 billion of national spending was $10 billion spent on iPhones. So this year when we calculate GDP we will calculate $90 billion-$0=$90 billion. GDP doesn’t change and that shouldn’t be surprising since GDP is about domestic production and the sinking of the container ship doesn’t change domestic production.

We continue:

If we want to understand the role of imports (and exports) on GDP and national welfare. We have to go beyond accounting to think about economics. If we permanently stopped all the container ships from delivering iPhones, for example, then domestic producers would start producing more cellphones and that would add to GDP but producing more cellphones would require producing less of other goods. If we were buying cellphones from abroad because producing them abroad requires fewer resources then GDP would actually fall—this is the standard argument for trade that you learned in your microeconomics class. The standard answer could change, however, if there were lots of unemployed resources, an issue we will discuss in Chapter 32 and later chapters. The point we want to emphasize here is not whether trade is ultimately good or bad but rather that Y+C+I+G+NX is an accounting identity that can’t by itself answer this question.


Macroeconomics = Tautologies + Daft Definitions.

Gross Domestic Cuckoldry is hitting all time highs these days.

Why would any of these letters be better at explaining themselves?

Cuckoldry is closer to being a science than economics.

I think it was the great Presidente Bolsonaro who admitted cuckholdry and became an incel eunuch in order to save his people from the anti-gun activists.

“[Sculpture is] a whole other state. And it’s a building thing. Whereas the painting is more fusing—fusing of ideas, fusing of feelings, fusing projected on atmosphere.”[23]

Y = C + I + G +X - M
Y = Cd +Cm + Id +Im +Gd +Gm +X - M
where d= domestic and m=imported. So, part of Consumption is imported, part consumed domestically. Part of Investment spending is composed by capital goods domestically bought, part is imported, and so on. And M = Cm+Im +Gm,
so Y = Cd +Id +Gd +X and imports should not impact Y.

The difficulty always seems to stem from trying to use an accounting identity as a model, without spelling out all of the assumptions underlying such an attempt.

Accounting identities how do they work? Well, apparently something like this - 'We have to go beyond accounting to think about economics.'

Leading to this sort of fantasy from the realm of economics - 'If we permanently stopped all the container ships from delivering iPhones, for example, then domestic producers would start producing more cellphones....' The U.S. has basically lost its ability to manufacture even the infrastructure to build cellphones, much less assemble cellphones using imported components, making that 'more' utterly delusional, along par with the hoary joke about economists just imagining whatever they require.

Well it is not difficult just a little tedious:

Neither the production approach nor the income approach even touch imports (only when calculating their taxes/subsidies). Only the expenditure approach, which this post explained, uses imports since it must be netted out or it will be double counted.

Probably just a typo...

I've noticed similar mistakes in the economist ever since it become "woke."

the more progressive its reporting has become the more sloppy its economic analysis has become.

i dont think its a coincidence. they probably hired token editors which is why such nonsense is printed.

The Economist became more "woke" in the early 1990s, that's when they lost their laisse faire edge and turned soft and socialist.

Wow, the definition of overreaction right here.

No, that's about right. I had a subscription back then when taking my econ classes, then afterwards.

Apart from a random article, I last read the Economist ten years ago. They were making errors all the time.

I stuck at it a little longer, but I stopped after they backed Obama over Mitt Romney in 2012. I just couldn't see the point any more: if the Economist can't be bothered to make the case for a capitalist as President, why read it? It's not as if there was a dearth of left wing papers telling us why we needed more Obama.

I sympathise. I could see the point of gambling on the unknown Obama when the alternative was the dreadful McCain. To prefer Obama, much better known, over Romney seemed foolish to me.

out of all the classification systems, economics has to be the most earnest. Just look at budgeting. After world war I, the ontological realization surpassed the existential question for the first time, or, there the epistemological concurrence in heaven disrupted intellectual concern on earth. There was no more reason to the think the expression of a tree was less important than a human.

The budget response was a reaction on how to measure presidential income as the president is the one person who can adjudicate on the elasticity of organ damage in the supply industry. The executive branch has tried to combat discrimination without poetic sexualization. This distortion can be a factor or function of subscription bases. In this way the presidential representation is a mimetic device that allows for correct sequencing in accounting affairs, bottom up…. A catch all idea from Austrian economics…

if you let the president budget the system as opposed to congress, because of polling, it is more democratic but less idiosyncratic.

Thre responsibility oof the executive branch is to untangle responsibility from nobles oblige. The responbility of provocateur is to harness the energy from mimicry, the provocateur is subordinated by his will, thus representative of the gift of sacrifice, the preponderance of serendipity while the treasurary secretary would insure the Huguenot. Assistant Secretary of the Navy Franklin Roosevelt also called for elliptical control of classification to denude the “je-nada” for the diamond’s purification. The French monopoly system it was charged by democrats, was being liquidated by the Russian oligarch.

How true!

In 1930, Babe Ruth was told he made more money ($80,000) than did the President Hoover ($75,000). His response, "Anyway, I had a better year than he did."

fun fact: articles in The Economist are not written by economists.

What about if consumption is defined as consumption of locally produced goods and services. Would this not make it clearer?

AlexT: "The point we want to emphasize here is not whether trade is ultimately good or bad but rather that Y+C+I+G+NX is an accounting identity that can’t by itself answer this question." - wrong. Free tade is always good, as it can be shown mathematically since the days of Ricardo. Whether it's unilateral trade (one country opening its borders while the rest of the world is closed, the 'traditional' USA stance and model), or bilateral trade (in theory what Trump is trying to do, with 'fair trade'), free trade is always good and increases the GDP/GNP of both countries engaged in trade.

Further, note AlexT's textbook gets it wrong in subtracting Imports in the first equation shown when in fact the accounting identity is like the second equation whown, which is *net* imports/exports (NX = X - I). What else does this $400 textbook have wrong?

$400? Was this considered in the last Mercatus piece on the high price of higher education?

"Free tade is always good, as it can be shown mathematically since the days of Ricardo."

It's wonderful that economists can mathematically show that things are good or bad.

Nope. You are imposing your own normative values on the social welfare function and imposing a homogeneity assumption. Even with homogeneity, if we value the utility of citizens more than the utility of foreigners, trade might be welfare reducing because some profits, wages, interest and rent are shipped abroad. Domestic consumers are unambiguously worse off with trade restrictions, and their great numbers suggest welfare losses. But this relies on homogeneity and equal value on everyone's utility.

For better or for worse, we put domestic profits and employment ahead of consumer welfare for decades. While I side with the consumers, that is not necessarily the case for the planner or the electorate.

Why not just re-order the components of the identity to avoid this far too common mistake? If C and I include imported goods M, then M should be on the opposite side of the equation from C and I, along with all their other components:

C + I + X = Y + M - G

What we can consume C, invest I, or sell to foreigners X consists of what we produce Y plus what we import M less whatever government siphons off G.

But what if government has a surplus in G? And can we ever have a negative Y? Why not? Y' (not) (compliment of Y! Not Y factorial either)

government has a surplus in G?

Isn't this why Greenspan testified at great length and greater confusion that he didn't know how to handle a surplus G? It was gone in a glimmer anyway,

I remember my first year in law school when the smartest students in my class, the engineering majors and the accounting majors, didn't do so well because the numbers didn't add up. Things improved for them in the second and third years when the numbers were more likely to add up. In economics it's the opposite: in the first year all of the numbers add up, but the longer one studies economics, the less likely the numbers add up.

if you need a garbadine handkerchief, we'll have to create a peace between the doorman and the elevator operator.

The answer is maybe.

There are no natural constraints to trade with Canada, so one would expect it takes many months to sort out what was made in Canada and what was made in the USA. Trucks can drive across as needed.

But a port of entry is constrained, queued and generally uses ship in both directions. In those cases the value of goods should tend to be proportional to the cost of shipment, and NX works as the shortest path. in fact, that is why we have import/export banks, because of constrained ports.

Easiest way to explain this is that if imports increase then C+I+G must increase by an exactly offsetting amount. So an import of a car by a consumer makes C go up then we need to subtract it off elsewhere via the -M because GDP is supposed to be measuring what we produced (in final goods value added etc etc) at home. So we don’t want to count the far. So it’s an accounting adjustment.

But I agree that economists confuse students by sometimes talking about GDP as a conceptual even theoretical device and sometimes as an accounting identity and the textbooks never explain clearly that there is this dual nature of the equation. No wonder students get confused.

Sorry, but you are wrong. Imports of final goods are not included in GDP, but Imported intermediate inputs are subtracted from GDP --- that is, they show up as actual negatives.

Here's the equation for GDP:

Y = C + I + G + (X − M)

Y is GDP
C is consumption
I is investment
G is government spending
X is exports
M is imports

It is pretty obvious that imports (M) is absolutely 100% part of the equation for GDP. Since there is a minus sign in front of it there's an inverse relationship to GDP (Y), so the Economist is not wrong here. I'm also not buying the Iphone example.

The problem is that C, I, and G are have a dependency on X and M because they represent total consumption, investment, and g-spending. If imports are down, then some amount of C, I, and G will be down too as you can't consume/invest in something that doesn't come to your borders as Alex/Tyler's example showed. Also, GDP doesn't measure what it says on the label. What does consumption or government spending have to do with production? Nothing. If the government overpays for a failed bridge to nowhere, that gets counted into GDP even though nothing was produced. GDP is a national spending measure with an odd net export adjustment. If you want to measure production you want something like return on investment/asset or profit/loss statements. Good luck changing this paradigm.

>If the government overpays for a failed bridge to nowhere, that gets counted into GDP even though nothing was produced

The failed bridge is a product that has been paid for.

It's no different from buying Reddit Gold; and bunch of meaningless one's and zero's on a remote computer that is accounted for in GDP.

It isn't an equation; it's an identity.

It's both. Odd that this is an econ blog but nobody seems to know basic things about economics. To be fair, GDP is not a particularly intuitive nor a good metric. Reminds me of the p-value controversy in the rest of the social sciences which is also neither particularly intuitive nor a good metric.

The p-value is perfectly informative. Its problem is that it surrenders the proper cost-benefit analysis of Type 1 and Type 2 errors to the reader. That's not necessarily a bad thing, especially since for any test statistic a reader can look up the p-value himself. But the practice of weighing the relative costs and benefits of statistical errors seems completely absent in academia. It is an intellectual punt.

A much more stupid and woke article in The Economist this week was titled "How to think about global warming and war". Typical bad editorializing uncritically pushing the alarmist idea that all things bad can be blamed on global warming. And then speculating on new ways to do so! Even more woke, a couple of weeks ago they endorsed voting for the ANC in S. Africa despite admitting their inferior policies and corruption, purely because of the racial composition of the parties. Uh, self-important young Economist journalist, S. African voters didn't need you to educate them on how to vote for their own race, did they?
My theory - journalists graduating from school are now consistently leftist. Only a heroic and expensive effort to find pro-market economically sophisticated journalists could buck that infiltration. But the current Economist editorial leadership is happily center-left, taking consensus views and trying to make them sound clever.

When Tyler interviewed Margaret Atwood a few weeks ago, she made the case that we still need feminism because of Global Warming. She said, Global Warming will cause War. And, War is bad for Women. So we still need Feminism.

I've thought about this from time to time since. I *think* Tyler was just leaving her with enough rope, because he didn't say anything in response.

Not long ago there was no such thing as a journalism school. Newspaper writers learned their craft (it's not a profession, more akin to laying bricks or painting houses) from tyrannical editors who had come up the same way. Page through any college newspaper and you'll see why "journalism" in the US has become a joke.

South Africa was a cause celebre throughout the west because it consisted of painless atonement for both colonialism and racism. Painless except for the whites that inhabited that unhappy land. At least the liberals of the west get to feel good about themselves without any loss of material goods or moral superiority. Their own minorities remain social inferiors.

Well, maybe.

Of course, if Country A buys less smartphones from Country , then money formerly exported is then spent on domestic producers, boosting GDP. Only if there is zero slack in the economy would this not apply.

So, when Japan imported less, the money not exported instead circulated domestically, boosting GDP.

The hope is plus minus not positive negative

The sinking ship example demonstrates a major weakness of GDP as a proxy for economic activity. We are unambiguously worse off when the ship of iPhones sinks, but GDP remains the same. Sure, the income is spent on other things, but other things of LESSER VALUE to consumers. Consumer surplus is lost with the ship, and possibly a fair amount of domestic producer surplus.

Macro is voodoo.

Accounting identities may be good way of summarizing information but they don’t explain anything. This point was made in a beautiful blog post from Nick Rowe

The guys at the Economist were just trying to "account" for the surprising rise in GDP, which is different from trying to explain it.

I think the Economist was right.

Consider that consumers in Japan buy shoes, some of which may be domestically produced and some of which are imported. Ignore inventory issues in this example for simplicity.

We were expecting based on, say, newspaper stories that Real GDP was going to be flat, but we find it rose instead. We will suppose the government statisticians got it right. GDP rose 2%, and we find that imports fell substantially instead of rising but consumer bought more shoes exactly as we forecasted.

The fall in imports does "account" for the surprising rise in GDP. What happened is that consumers bought more shoes from domestic producers and fewer from foreigners. That means that domestic shoe production (GDP) must have been higher than we expected. Our estimate for imports has direct implications for our estimate of domestic production. We may not be able to measure domestic shoe production directly, but we can back it out from learning about consumer purchases of shoes and imports of shoes. The Economist staff understands this.

But we still need an explanation for the surprise in imports? Did the exchange change, pushing the prices of imported shoes up, so consumers substituted into domestic shoes? Did the government lower the quota of imported shoes? Was there a strike in foreign shoe factories, cutting off supply? Let us debate the explanations for what happened to GDP, but let us not argue about simple accounting identities.

Good response. I'm still trying to digest it to determine whether I agree.

Accounts for, not explains. Plausible.

But wouldn't you agree that the casual reader less versed in the nuances of GDP accounting might view this favorably without regard to whether it was a fluke or, perchance, the result of something dreadful for them personally?

When governments and economists report on GDP in this fashion, are they really saying anything or merely making throat noises?

I agree that one cannot use the identity to state that "in order to increase GDP a country should reduce imports."

However, in the limited sense used in the article, to explain an unusual result in a single quarter, this may make sense. In the iPhone container example, it is as if the container arrived in one quarter (depressing GDP) and the consumption happened in the next (boosting GDP). Overall, the effect is zero, but there is a timing effect.

But then that sort of reporting would be malinformative to all but the most patient and astute observers.

The reporters are simply making noises resembling good grammar and meticulous data. It would be like judging the performance of a stock based on a single tick.

Comments for this post are closed