Great article, but the writer buried the most persuasive parts at the end. He could have swayed more minds by leading with his point that “Tariffs on intermediate goods generally hurt GDP more than tariffs on final goods” and then making the case from there. Just my two cents. Either way, thank you for sharing your expertise on the subject with us.
IMHO the key insight (other than hexapodia) is, that we're trying to use measures of _spending_ to infer a measure of _production_.
If it were the case that everything consumed in the US had to be produced in the US, then spending would necessarily be equivalent to the sale value of everything that was produced-and-sold. (Conceivably there could still be a small wedge between production and purchases, in the form of changes in the level of inventory.)
But then because we have exports and imports, if we want to use the spending numbers to try to infer the value of production, we have to adjust for them, by adding in the spending of foreigners on domestic production (exports), and subtracting out the spending of residents on foreign goods (imports). Because foreign exchange transactions are involved, these are also (relatively) easy to track.
We of course also do try to directly estimate value of production with various surveys of producers, but I think the consensus is that the spending estimate is more accurate. (e.g. Because the GDP number can spike or crash depending on the fortunes of small businesses. In a boom, lots of small new companies form and do very well. New businesses are slow to enter the producer survey data -- the surveyors may not even know they exist during their first couple years, and their response rate on surveys, even if they get identified, is worse.)
Or course imports have to be subtracted from our consumption when measuring consumption in GDP.
I suppose one can say they aren’t in GDP after that subtraction, but that is just arithmetic, and doesn’t really add anything to the debate about whether more imports or more exports are good or bad. It depends, as you note. Moreover, the flip side of negative current account balances are capital account surpluses (loans and investments from foreigners), as you know, and the current fad in econ is to consider these the driver of net exports. That, too, is a circular argument as one could say that consumption suppression and protectionism (and sometimes currency manipulation) lead to positive X-M in mercantilist economies (Germany, China, Japan) and high savings rates (ie reducing imports and increasing exports IS the driver), but it all zeroes out after the fact (the flows leading up the zeroing out are importantly).
My view is that the US imports and consumes too much while Germany and China consume too little. Are you arguing the opposite? That is the real policy debate. Reducing US imports, reducing debt-fueled consumption (7 pct deficits at full employment - insanity) and increasing consumption in Europe and Asia would be desirable. Then we can talk about policies. Tariffs are just one of those policies (not one I would choose). Moreover, should the US stock and bond markets be soaking up so much of global flows when investment is needed in Africa, West and South Asia and Latam? I note that China for more than a decade now has been investing its surpluses in those regions (more to buy influence and secure nat resources than on the basis of profit or development objectives). Unfortunately some of that money ends up straight back in dollars in the Swiss Bank accounts of these nations’ potentates and officials.
I mean, it is great that I can get cheap Happy Meal toys and my house value keeps rising and stocks can trade at 10x sales, but just because the US (as the reserve currency) can borrow to consume, does it mean we should? And is the debate about whether the US in importing too much, offshoring too much, consuming and borrowing too much really about X-M and GDP- of course not.
Yes, that is the real policy debate - BUT NOT AT all what Noah was addressing here... He was simply addressing why the fallacy (and one that has big impacts) is widely held and what is wrong with it....
I'd say that a considerable part of the argument is motivated reasoning, but it's not incorrect when you realize that the motivation is "increase the income of US manufacturing workers *relative to* the income of US 'professional class' workers". There's an underlying problem in economics in that it assumes the goal is to maximize GDP-per-capita, or consumption, or something similar. But the actual goal of humans is "maximize my status position in society". Now when you're buying toilet paper, or some other transaction among a small number of people, these turn out to be nearly the same -- you want to make money to raise yourself versus the masses but you don't care if the three other people involved are raised up or how much, because they are statistically insignificant relative to the masses. But when a political choice decreases the consumption of your class by $100 each and decreases the consumption of the other class by $200 each, *your status improves*. That's why laid-off manufacturing workers want protective tariffs. Of course, it would be cheaper to just institute an official income redistribution program, but that would have the stigma of welfare, and a big status distinction between the "working class" and the class immediately below it, the "poor", is that the working class aren't "on welfare".
Having disabused Trumpists of their errors about tariffs, would you also do the same for Progressives about why government purchases, or subsidies for private purchases, of say new solar power generation or roof-top solar do not "create jobs."
The question of whether government spending creates jobs is dependent on whether you're already at full employment (i.e. unemployment <= NAIRU). If you're in a severe recession, spending absolutely will raise employment -- you're mobilizing resources that were unnecessarily sitting on the sidelines because of coordination problems.
If you're in a boom, it will simply draw people from one industry to another, and cause inflation both in the subsidized industry, and in whatever industries are being forced to match the subsidized industry's new power to bid up the prices of wages and material inputs, so those competing industries can continue operating even at reduced output.
This is Macro 101 stuff, and insisting that government spending _never_ creates jobs is just as silly as insisting that it _always_ creates jobs.
I think it's not so much that progressives claim that such purchases "create jobs" than progressives argue against conservatives position that such purchases destroy jobs.
While the conservatives are correct that jobs will be lost the progressive position that other jobs will be created is also true. As Treeamigo says in this thread, there is a substitution effect going on. Government is putting its finger on the scale to move jobs from what progressives (and I) consider less desirable industries (e.g. oil, coal extraction) to more desirable industries like solar power.
The problem is that both sides are doing it wrong. The right answer is to do both. If we can increase oil production by 1 mbbl/day and also reduce domestic oil demand by 1 mbbl/day (by investing in renewables) that creates more jobs in both industries and also an additional 2 mbbl/day of oil to export for an additional $50 billion / year.
Bingo. Conservatives demonize subsidies to Renewables; but either laud or are blissfully unaware of our much longer history of subsidies to the Fossil Fuel industry.
Because FOX News either misinforms or fails to properly inform them.
How is it that you think that spending money doesn’t create jobs? Have you gone communist on us? Or are you just stating the obvious, that despite creating jobs in “roof top solar,” it has now failed to create jobs in all the other endeavors where it might have been spent?
If the government gives me a 10 pct rebate on chicken, maybe I buy more chicken than pork. That is substitution. Then they tax my children to pay for all of that chicken. That is progressivism.
"Imports have nothing to do with domestic economic production. The number of asteroid impacts in the Andromeda galaxy is doesn’t count in U.S. GDP either."
While neither are "included" in GDP, the former is vastly more relevant to GDP than the latter. It's also accurate to say that trade balance can and does affect living standards (which is really what Navarro and Trump are talking about.) The classic extreme case of this cycle is Nauru.
Consumer welfare myopia has plagued economics since Adam Smith, but I see economists more willing to consider production these days. I welcome this. Ian Fletcher, Gomory and Baumol all say that for an advanced economy like the United States, a flat tariff can make sense. America is vast, resource rich, educated, and industrialized; we have comparative advantages in a great majority of products. A flat tariff does make Peruvian bananas more expensive, but it also makes American made everything more competitive (especially domestically). Even the banana example (absurd as it appears) works this way, likely raising the demand for Hawaiian pineapples slightly. (Not saying this is worth it, just pointing out the likely substitution effects.) Targeted tariffs also have a significant disadvantage over universal ones: they are easily prone to regulatory capture by the concentrated interests who profit by them. (And yes, I do realize that a corresponding tariff on our exports would likely be levied, but again, this is a place where, a large, diverse economy has a distinct advantage.)
I think both Harris and Trump's economic "plans" are pretty idiotic and grossly inflationary (hers mildly; his insanely, as you've previously said), but on this point, Trump's team has more basis than you give them credit for.
This is exactly the same mistake people make when they think that riding the subway reduces traffic or that riding your bike or an electric car reduces emissions.
No.
Each of those things fails to produce traffic or fails to produce emissions, just as imports fail to produce gdp. But they only count as a reduction if you are implicitly assuming a baseline where you do exactly the same travel or exactly the same consumption, and assuming a default that everyone drives a gasoline car or that everyone consumes domestic products.
There are some contexts in which this is reasonable - someone who used to commute daily in a gasoline car who switches to a bike is reducing traffic and emissions, and someone who used to buy us made steel and switches to Canadian made steel is reducing us steel production. But plenty of trips on bike or subway, and plenty of imports, are consumption that just wouldn’t have happened without the alternative being available, and thus can’t reasonably be counted as reductions. Your examples of coffee and bananas make that clear.
What are the emissions of the cows used to generate the 1500 calories that person will burn on their 60 minute commute each day? (I'm actually not being facetious; I really wonder.)
I do think that’s an important question. This is the first result I found when searching:
> When considering the GHG savings of cycling, it is key to make accurate estimates of the carbon emissions associated with the activity. From the estimates above, we can see that cycling actually has non-negligible GHG contributions, largely due to the emissions caused by the agricultural sector in producing food and raising livestock. At the extremes, a vegan cyclist will produce only 5% of the emissions a conventional pickup truck will produce, while a meat-loving cyclist will actually produce 42% more GHGs than the most efficient EV. For the more ordinary case, the average cyclist's GHG contribution (~60 g CO2e km-1) is nearly one quarter that of the average car (~250 g CO2e km-1). Thus, for the average case, switching from driving to cycling decreases the carbon footprint of the trip by 75%.
Funny. Has never occurred to me that people believe in tariffs because of that equation. Maybe we should have special national accounts that net out imports from the demand components. I always thought people believe in tariffs because they believe that the import supply elasticity is very high relative to the import demand elasticity.
My favorite Balance of Payments fallacy is that politicians want current account superavits and net foreign investment in their country. This makes complete sense, because both things create employment.
Unfortunately it is an accounting (almost arithmetic!) imposibility (because your current account superávit is equal to your net investment abroad).
This happens all the time in politics: while people understand their local interests, the general equilibrium consequences are always anti intuitive.
The best case is “polygamy is good for man: if you don’t want more than one woman, don’t marry more”; of course with similar male and female population, the equilibrium consequence of polygamy is a deficit of women. “Solve for the equilibrium” as T Cowen says…
China’s solicitation of foreign investment and its desire for the Yuan to become a reserve currency while at the same time running a mercantilist export-focused economy does seem to be arithmetic-challenged.
Then again, the current fad is to believe that capital flows drive trade flows, so by encouraging foreign investment they would be reducing net exports at the margin.
Of course, in any case, there is not relation between which country you have current account deficit/superávit and bilateral investment, an additional complication for politicians
I think polygamy usually involves older men marrying younger women; and in a high-birth-rate population, there will be more younger people than older people. So the "similar male and female population" assumption doesn't necessarily apply.
I think Ran is assuming a high rate of attrition of the young men, before they become old men. If you have a culture that sends lots of men into extremely destructive wars, or puts them into high-mortality professions, that could in theory create an equilibrium, where the average 40 year old man who is now a manager or officer, rather than a frontline worker, can have at least one 20 year old wife, and the luckier managers and officers can have two. Say mortality is comparable for men and women for ages 0-20, but upon entering the working world, 30% of men who made it to age 20 die before they reach age 40. So with a stable population overall, 30% of the 40-year-old men can get an extra 20-year-old wife. If the average birth rate per woman is high enough that population is growing over time despite the high mortality rate (after all, somebody needs to populate the conquest frontier created by the war machine that's killing all those men), there will be even a slightly larger surplus of 20-year-old women.
No modern society is like this, though, and IMHO it would be a moral atrocity to try to make one be like this. Certainly there's no way that _just_ population growth could suffice. In the absence of some driver of extreme mortality between ages 20 and 40, population growth at 2% per generation would mean only 2% of the 40 year old men could get an "extra" wife. If population were doubling every generation, sure, the 40 year old men could all get an extra 20 year old wife, but that obviously is unsustainable in a world where basically all habitable areas are already inhabited.
> I think Ran is assuming a high rate of attrition of the young men, before they become old men.
Not at all!
> with a stable population overall
Why are you assuming a stable population?? It's completely normal -- in fact, it's the norm -- for populations to grow over time. So even if the gender ratio is exactly 1:1 and literally no one dies under the age of 80, there will be more 20-year-old women than 35-year-old men at any given time, because more babies will have been born in year x-minus-20 than in year x-minus-35. And the faster the population grows, the greater the disparity will be.
Does that make sense?
> No modern society is like this, though, and IMHO it would be a moral atrocity to try to make one be like this.
To be clear: I'm not advocating polygamy! The argument that Arturo cites is explicitly only about polygamy as a benefit for men -- it literally does not consider the desires of women at all. That's a moral atrocity no matter what other assumptions we might make!
A modern society cannot be "high birth rate" without either also having a high mortality rate, or having its population _explode_.
I'm not at all a "limits to growth" fan. (If anything, I'm an Yglesias "One Billion Americans" fan.) But it's impossible to imagine a society that, over the course of multiple generations, could allow more than the _tiniest_ sliver of 40-year-old men (like, on the order of 2%, not 20%) to have more-than-one 20-year-old wife, without either leaving a bunch of 40-year-old men unmarried, or having a lot of the men die before they reach 40. If one of those isn't true, then the population growth rate is so fast that you actually _would_ be looking at problems with pushing up against Malthusian limits, in short order.
All correct [that phrase apparently being the origin of "OK" :)], but I wish you had also gone through the "macro" part of the explanation. Tariffs which lower imports "strengthen" the dollar and reducing exports and increasing imports of anything imported that does NOT have a traffic.
This is important becasue it also teaches the importance of the the exchange rate effect which in turn is important for understanding how fiscal deficits ALSO reduce exports and increase imports. And this is the mechanism -- a shift in the relative prices of tradable goods and services like midwestern agriculture and manufacturing -- that I think was largely at work in the "China Shock."
Thomas, considering we are the world's reserve currency though, do you think this effect is actually visible in the USD? I'm not sure, but you're a far better macro guy than I am.
(Yes, there's a point to the story below...) Airplane autopilot systems used to have problems in icing conditions. As ice builds up on the wings, the lift becomes choppy and rough. The autopilot can correct for this for a long time, and it does it quickly enough that (at least for a while) the effect is not noticeable. But eventually, the ice buildup gets large enough that the autopilot craps out. Then the pilot is suddenly handed a plane whose lift profile is so degraded that no human could possibly control it. My understanding is that this did actually cause accidents until icing sensors were added.
I wonder if our reserve currency status is functioning in the same way, insulating our currency from the corrective effects of an obviously out-of-whack trade balance. I'm honestly not sure, but it's a high-risk enough event that I think it's worth considering. What happened to Liz Truss gives an idea how quickly currencies can go south.
You have spent a great deal of effort attempting to correct a fundamentally flawed statistic i.e. GDP.
1. Markets exist for the single purpose of satisfying consumer wants and needs. Money spent in the other 3 (or 4) components ultimately ends up spent on consumption. (Check Say's Law.)
2. Specifically, government "spending" must come out of consumption or investment (part of #1.)
3. None of the components of GDP have any significant meaning in the aggregate. (How do you add dollars spent on Big Macs to dollars spent on iPhones and derive any significant conclusion.)
Like most macroeconomic data, GDP only provides employment for statisticians and economists. It does not really help consumers. producers, or investors.
Of course, a lot of the discussion here assumes GDP measures something meaningful (not actually the point of Noah's post which is about a common fallacy).
The classic example being a tornado that destroys a town. GDP goes up with the rebuilding effort. but everyone would be hard pressed to say that the country is better off overall as serious wealth has been destroyed and probably not recovered at best.
However, since we don't have a better measure that is commonly accepted of "general welfare/wealth", I guess GDP will have to do. But again, like many other commonly used terms, most Americans think it implies increase in overall wealth or welfare and that isn't what it measures.
A long explanation of something so simple to understand from its name that the people who are confused don’t know how to read. GDP: Gross Domestic Product. Obviously to anyone who passed 5th grade reading comprehension it refers to what was produced domestically. If there are creatures who can’t read at the 5th grade level only a moron would take advice from them.
“Let’s talk about what GDP is. GDP is the total value of everything produced in a country:
GDP = all the stuff we produce.”
101 question, value is different from what is produced and is not a fixed number like the what is produced. 100 widgets selling for $10 each has a different value if the price goes up or down due to interest or availability in purchasing the widget. However, the number of widgets produced remains the same at 100 despite changes in value. So value as a variable can’t be equal to production which is a fixed number and not variable. What am I missing?
Noah is using “production” as shorthand for “value of production”.
In international comparisons they usually don’t want one country to count as having artificially high or low gdp because some goods are sold at a different price in the different countries, so they correct for purchasing power parity. Similarly, they try to correct for inflation across years. I think both of these measures try to differentiate change in price caused by change in quality of goods from mere change in price of the same goods but I don’t know how well they do that.
Thanks Kenny and I may be too thick or perhaps too stubborn and would be better off not pushing for clarity I may not be committed to enough to produce (create value for myself).
With your input Noah is saying the same thing twice and simply dropping the word value from the second iteration. Ostensibly, this is for clarity.
It seems I’m unclear about what value means. It seems elusive as there is value from the seller’s view and perhaps a different value from the buyer view with the ultimate value being an agreement reached between the two.
If GDP is the value of all production and value is a varying amount based on agreement, it seems GDP ends up being fundamentally the aggregate agreements about value and total production is just a multiplier of that agreement.
It makes economics seem like a swishy practice pretending to be more solid than it actually is.
I think that most economists think GDP matters mainly because it is a measurement of how much human welfare is created by all the activities going on - higher GDP means more of what people care about is going on. Figuring out how to translate market prices into what people actually care about is inherently difficult, and I think that economists who are at all philosophically inclined will agree with all of that.
But this sort of conceptual complexity is more common across all the sciences than people often imagine. Economics probably has more difficulties, because it’s trying to measure what matters to people, rather than measuring some physical thing that isn’t directly what people care about but is easier to measure. But because it is about what matters, I do think it’s still important to try.
Yes, anything can be measured if you think it can and agree on a measure. Yes, any measure ends up bound by language as it’s a product of “languaging.”
Your “but is easier to measure” recalls the story of a man on his hands and knees under a light at night in a parking lot looking for his keys. Another man joins in and after a while feeling exasperated asks the first man “Well where did you lose the keys???” The man pointed to the far unlit corner of the lot and said “I think over there but there’s very little light there.” Most anything valuable begins by being truthful.
Sorry I wasn't clear. I’m not making a claim, I’m asking a question seeking clarification. The writer is defining GDP by stating what it is. First he says it is value then he says it is equal to production. I of course do not know how is and equal are being considered by the writer or if he means to be that precise in his usage of either term when combining them in the same sentence. It’s just unclear to me at this point.
thanks so much for the clear explanation. Yet the line between what the NYT wrote, and what the BEA reports, is subtle: "... primarily reflected upward revisions to private inventory investment and federal government spending that were offset by downward revisions to nonresidential fixed investment and exports (refer to "Updates to GDP"). Imports, which are a subtraction in the calculation of GDP, were revised up." https://www.bea.gov/index.php/news/2024/gross-domestic-product-third-estimate-corporate-profits-revised-estimate-and-gdp-0
Very late to the party but isn't the purpose of minusing exports also relevant to exports too? Given the fact that you can import x, perform some value-add to x and then export that new product. Countries don't record the entire value of the export as GDP, just the value add. Otherwise you wouldn't see countries with export as a % of GDP greater than 100
Great article, but the writer buried the most persuasive parts at the end. He could have swayed more minds by leading with his point that “Tariffs on intermediate goods generally hurt GDP more than tariffs on final goods” and then making the case from there. Just my two cents. Either way, thank you for sharing your expertise on the subject with us.
A follow-up post will focus on this. :-)
It might be easier to understand the "formula" as follows:
GDP = (Consumption + Investment + Government purchases - Imports) + Exports
That's pretty good, yeah.
IMHO the key insight (other than hexapodia) is, that we're trying to use measures of _spending_ to infer a measure of _production_.
If it were the case that everything consumed in the US had to be produced in the US, then spending would necessarily be equivalent to the sale value of everything that was produced-and-sold. (Conceivably there could still be a small wedge between production and purchases, in the form of changes in the level of inventory.)
But then because we have exports and imports, if we want to use the spending numbers to try to infer the value of production, we have to adjust for them, by adding in the spending of foreigners on domestic production (exports), and subtracting out the spending of residents on foreign goods (imports). Because foreign exchange transactions are involved, these are also (relatively) easy to track.
We of course also do try to directly estimate value of production with various surveys of producers, but I think the consensus is that the spending estimate is more accurate. (e.g. Because the GDP number can spike or crash depending on the fortunes of small businesses. In a boom, lots of small new companies form and do very well. New businesses are slow to enter the producer survey data -- the surveyors may not even know they exist during their first couple years, and their response rate on surveys, even if they get identified, is worse.)
Or course imports have to be subtracted from our consumption when measuring consumption in GDP.
I suppose one can say they aren’t in GDP after that subtraction, but that is just arithmetic, and doesn’t really add anything to the debate about whether more imports or more exports are good or bad. It depends, as you note. Moreover, the flip side of negative current account balances are capital account surpluses (loans and investments from foreigners), as you know, and the current fad in econ is to consider these the driver of net exports. That, too, is a circular argument as one could say that consumption suppression and protectionism (and sometimes currency manipulation) lead to positive X-M in mercantilist economies (Germany, China, Japan) and high savings rates (ie reducing imports and increasing exports IS the driver), but it all zeroes out after the fact (the flows leading up the zeroing out are importantly).
My view is that the US imports and consumes too much while Germany and China consume too little. Are you arguing the opposite? That is the real policy debate. Reducing US imports, reducing debt-fueled consumption (7 pct deficits at full employment - insanity) and increasing consumption in Europe and Asia would be desirable. Then we can talk about policies. Tariffs are just one of those policies (not one I would choose). Moreover, should the US stock and bond markets be soaking up so much of global flows when investment is needed in Africa, West and South Asia and Latam? I note that China for more than a decade now has been investing its surpluses in those regions (more to buy influence and secure nat resources than on the basis of profit or development objectives). Unfortunately some of that money ends up straight back in dollars in the Swiss Bank accounts of these nations’ potentates and officials.
I mean, it is great that I can get cheap Happy Meal toys and my house value keeps rising and stocks can trade at 10x sales, but just because the US (as the reserve currency) can borrow to consume, does it mean we should? And is the debate about whether the US in importing too much, offshoring too much, consuming and borrowing too much really about X-M and GDP- of course not.
Yes, that is the real policy debate - BUT NOT AT all what Noah was addressing here... He was simply addressing why the fallacy (and one that has big impacts) is widely held and what is wrong with it....
I'd say that a considerable part of the argument is motivated reasoning, but it's not incorrect when you realize that the motivation is "increase the income of US manufacturing workers *relative to* the income of US 'professional class' workers". There's an underlying problem in economics in that it assumes the goal is to maximize GDP-per-capita, or consumption, or something similar. But the actual goal of humans is "maximize my status position in society". Now when you're buying toilet paper, or some other transaction among a small number of people, these turn out to be nearly the same -- you want to make money to raise yourself versus the masses but you don't care if the three other people involved are raised up or how much, because they are statistically insignificant relative to the masses. But when a political choice decreases the consumption of your class by $100 each and decreases the consumption of the other class by $200 each, *your status improves*. That's why laid-off manufacturing workers want protective tariffs. Of course, it would be cheaper to just institute an official income redistribution program, but that would have the stigma of welfare, and a big status distinction between the "working class" and the class immediately below it, the "poor", is that the working class aren't "on welfare".
I demand a windfall profits tax on toilet paper retro-active to March of 2020!
Sorry, I just couldn't resist.
Overall, I think your point is a very good one. There's vastly more to this picture than per-capita GDP.
Having disabused Trumpists of their errors about tariffs, would you also do the same for Progressives about why government purchases, or subsidies for private purchases, of say new solar power generation or roof-top solar do not "create jobs."
The question of whether government spending creates jobs is dependent on whether you're already at full employment (i.e. unemployment <= NAIRU). If you're in a severe recession, spending absolutely will raise employment -- you're mobilizing resources that were unnecessarily sitting on the sidelines because of coordination problems.
If you're in a boom, it will simply draw people from one industry to another, and cause inflation both in the subsidized industry, and in whatever industries are being forced to match the subsidized industry's new power to bid up the prices of wages and material inputs, so those competing industries can continue operating even at reduced output.
This is Macro 101 stuff, and insisting that government spending _never_ creates jobs is just as silly as insisting that it _always_ creates jobs.
I think it's not so much that progressives claim that such purchases "create jobs" than progressives argue against conservatives position that such purchases destroy jobs.
While the conservatives are correct that jobs will be lost the progressive position that other jobs will be created is also true. As Treeamigo says in this thread, there is a substitution effect going on. Government is putting its finger on the scale to move jobs from what progressives (and I) consider less desirable industries (e.g. oil, coal extraction) to more desirable industries like solar power.
The problem is that both sides are doing it wrong. The right answer is to do both. If we can increase oil production by 1 mbbl/day and also reduce domestic oil demand by 1 mbbl/day (by investing in renewables) that creates more jobs in both industries and also an additional 2 mbbl/day of oil to export for an additional $50 billion / year.
Bingo. Conservatives demonize subsidies to Renewables; but either laud or are blissfully unaware of our much longer history of subsidies to the Fossil Fuel industry.
Because FOX News either misinforms or fails to properly inform them.
How is it that you think that spending money doesn’t create jobs? Have you gone communist on us? Or are you just stating the obvious, that despite creating jobs in “roof top solar,” it has now failed to create jobs in all the other endeavors where it might have been spent?
If the government gives me a 10 pct rebate on chicken, maybe I buy more chicken than pork. That is substitution. Then they tax my children to pay for all of that chicken. That is progressivism.
"Imports have nothing to do with domestic economic production. The number of asteroid impacts in the Andromeda galaxy is doesn’t count in U.S. GDP either."
While neither are "included" in GDP, the former is vastly more relevant to GDP than the latter. It's also accurate to say that trade balance can and does affect living standards (which is really what Navarro and Trump are talking about.) The classic extreme case of this cycle is Nauru.
Consumer welfare myopia has plagued economics since Adam Smith, but I see economists more willing to consider production these days. I welcome this. Ian Fletcher, Gomory and Baumol all say that for an advanced economy like the United States, a flat tariff can make sense. America is vast, resource rich, educated, and industrialized; we have comparative advantages in a great majority of products. A flat tariff does make Peruvian bananas more expensive, but it also makes American made everything more competitive (especially domestically). Even the banana example (absurd as it appears) works this way, likely raising the demand for Hawaiian pineapples slightly. (Not saying this is worth it, just pointing out the likely substitution effects.) Targeted tariffs also have a significant disadvantage over universal ones: they are easily prone to regulatory capture by the concentrated interests who profit by them. (And yes, I do realize that a corresponding tariff on our exports would likely be levied, but again, this is a place where, a large, diverse economy has a distinct advantage.)
I think both Harris and Trump's economic "plans" are pretty idiotic and grossly inflationary (hers mildly; his insanely, as you've previously said), but on this point, Trump's team has more basis than you give them credit for.
This is exactly the same mistake people make when they think that riding the subway reduces traffic or that riding your bike or an electric car reduces emissions.
No.
Each of those things fails to produce traffic or fails to produce emissions, just as imports fail to produce gdp. But they only count as a reduction if you are implicitly assuming a baseline where you do exactly the same travel or exactly the same consumption, and assuming a default that everyone drives a gasoline car or that everyone consumes domestic products.
There are some contexts in which this is reasonable - someone who used to commute daily in a gasoline car who switches to a bike is reducing traffic and emissions, and someone who used to buy us made steel and switches to Canadian made steel is reducing us steel production. But plenty of trips on bike or subway, and plenty of imports, are consumption that just wouldn’t have happened without the alternative being available, and thus can’t reasonably be counted as reductions. Your examples of coffee and bananas make that clear.
What are the emissions of the cows used to generate the 1500 calories that person will burn on their 60 minute commute each day? (I'm actually not being facetious; I really wonder.)
I do think that’s an important question. This is the first result I found when searching:
> When considering the GHG savings of cycling, it is key to make accurate estimates of the carbon emissions associated with the activity. From the estimates above, we can see that cycling actually has non-negligible GHG contributions, largely due to the emissions caused by the agricultural sector in producing food and raising livestock. At the extremes, a vegan cyclist will produce only 5% of the emissions a conventional pickup truck will produce, while a meat-loving cyclist will actually produce 42% more GHGs than the most efficient EV. For the more ordinary case, the average cyclist's GHG contribution (~60 g CO2e km-1) is nearly one quarter that of the average car (~250 g CO2e km-1). Thus, for the average case, switching from driving to cycling decreases the carbon footprint of the trip by 75%.
http://large.stanford.edu/courses/2022/ph240/schutt2/
Funny. Has never occurred to me that people believe in tariffs because of that equation. Maybe we should have special national accounts that net out imports from the demand components. I always thought people believe in tariffs because they believe that the import supply elasticity is very high relative to the import demand elasticity.
Actually, I meant “very low.”
My favorite Balance of Payments fallacy is that politicians want current account superavits and net foreign investment in their country. This makes complete sense, because both things create employment.
Unfortunately it is an accounting (almost arithmetic!) imposibility (because your current account superávit is equal to your net investment abroad).
This happens all the time in politics: while people understand their local interests, the general equilibrium consequences are always anti intuitive.
The best case is “polygamy is good for man: if you don’t want more than one woman, don’t marry more”; of course with similar male and female population, the equilibrium consequence of polygamy is a deficit of women. “Solve for the equilibrium” as T Cowen says…
China’s solicitation of foreign investment and its desire for the Yuan to become a reserve currency while at the same time running a mercantilist export-focused economy does seem to be arithmetic-challenged.
Then again, the current fad is to believe that capital flows drive trade flows, so by encouraging foreign investment they would be reducing net exports at the margin.
Of course, in any case, there is not relation between which country you have current account deficit/superávit and bilateral investment, an additional complication for politicians
I think polygamy usually involves older men marrying younger women; and in a high-birth-rate population, there will be more younger people than older people. So the "similar male and female population" assumption doesn't necessarily apply.
I would say that in principle that aggravates the problem, doesn’t it? Do many old women marry with excess young men?
I think Ran is assuming a high rate of attrition of the young men, before they become old men. If you have a culture that sends lots of men into extremely destructive wars, or puts them into high-mortality professions, that could in theory create an equilibrium, where the average 40 year old man who is now a manager or officer, rather than a frontline worker, can have at least one 20 year old wife, and the luckier managers and officers can have two. Say mortality is comparable for men and women for ages 0-20, but upon entering the working world, 30% of men who made it to age 20 die before they reach age 40. So with a stable population overall, 30% of the 40-year-old men can get an extra 20-year-old wife. If the average birth rate per woman is high enough that population is growing over time despite the high mortality rate (after all, somebody needs to populate the conquest frontier created by the war machine that's killing all those men), there will be even a slightly larger surplus of 20-year-old women.
No modern society is like this, though, and IMHO it would be a moral atrocity to try to make one be like this. Certainly there's no way that _just_ population growth could suffice. In the absence of some driver of extreme mortality between ages 20 and 40, population growth at 2% per generation would mean only 2% of the 40 year old men could get an "extra" wife. If population were doubling every generation, sure, the 40 year old men could all get an extra 20 year old wife, but that obviously is unsustainable in a world where basically all habitable areas are already inhabited.
True. Excess men can be killed instead of being left as excess alive men….
That's what World Wars are for. To cull younger men.
> I think Ran is assuming a high rate of attrition of the young men, before they become old men.
Not at all!
> with a stable population overall
Why are you assuming a stable population?? It's completely normal -- in fact, it's the norm -- for populations to grow over time. So even if the gender ratio is exactly 1:1 and literally no one dies under the age of 80, there will be more 20-year-old women than 35-year-old men at any given time, because more babies will have been born in year x-minus-20 than in year x-minus-35. And the faster the population grows, the greater the disparity will be.
Does that make sense?
> No modern society is like this, though, and IMHO it would be a moral atrocity to try to make one be like this.
To be clear: I'm not advocating polygamy! The argument that Arturo cites is explicitly only about polygamy as a benefit for men -- it literally does not consider the desires of women at all. That's a moral atrocity no matter what other assumptions we might make!
A modern society cannot be "high birth rate" without either also having a high mortality rate, or having its population _explode_.
I'm not at all a "limits to growth" fan. (If anything, I'm an Yglesias "One Billion Americans" fan.) But it's impossible to imagine a society that, over the course of multiple generations, could allow more than the _tiniest_ sliver of 40-year-old men (like, on the order of 2%, not 20%) to have more-than-one 20-year-old wife, without either leaving a bunch of 40-year-old men unmarried, or having a lot of the men die before they reach 40. If one of those isn't true, then the population growth rate is so fast that you actually _would_ be looking at problems with pushing up against Malthusian limits, in short order.
All correct [that phrase apparently being the origin of "OK" :)], but I wish you had also gone through the "macro" part of the explanation. Tariffs which lower imports "strengthen" the dollar and reducing exports and increasing imports of anything imported that does NOT have a traffic.
This is important becasue it also teaches the importance of the the exchange rate effect which in turn is important for understanding how fiscal deficits ALSO reduce exports and increase imports. And this is the mechanism -- a shift in the relative prices of tradable goods and services like midwestern agriculture and manufacturing -- that I think was largely at work in the "China Shock."
Thomas, considering we are the world's reserve currency though, do you think this effect is actually visible in the USD? I'm not sure, but you're a far better macro guy than I am.
(Yes, there's a point to the story below...) Airplane autopilot systems used to have problems in icing conditions. As ice builds up on the wings, the lift becomes choppy and rough. The autopilot can correct for this for a long time, and it does it quickly enough that (at least for a while) the effect is not noticeable. But eventually, the ice buildup gets large enough that the autopilot craps out. Then the pilot is suddenly handed a plane whose lift profile is so degraded that no human could possibly control it. My understanding is that this did actually cause accidents until icing sensors were added.
I wonder if our reserve currency status is functioning in the same way, insulating our currency from the corrective effects of an obviously out-of-whack trade balance. I'm honestly not sure, but it's a high-risk enough event that I think it's worth considering. What happened to Liz Truss gives an idea how quickly currencies can go south.
We'll find out for real in a few years when the Yuan supplants the dollar.
You have spent a great deal of effort attempting to correct a fundamentally flawed statistic i.e. GDP.
1. Markets exist for the single purpose of satisfying consumer wants and needs. Money spent in the other 3 (or 4) components ultimately ends up spent on consumption. (Check Say's Law.)
2. Specifically, government "spending" must come out of consumption or investment (part of #1.)
3. None of the components of GDP have any significant meaning in the aggregate. (How do you add dollars spent on Big Macs to dollars spent on iPhones and derive any significant conclusion.)
Like most macroeconomic data, GDP only provides employment for statisticians and economists. It does not really help consumers. producers, or investors.
Of course, a lot of the discussion here assumes GDP measures something meaningful (not actually the point of Noah's post which is about a common fallacy).
The classic example being a tornado that destroys a town. GDP goes up with the rebuilding effort. but everyone would be hard pressed to say that the country is better off overall as serious wealth has been destroyed and probably not recovered at best.
However, since we don't have a better measure that is commonly accepted of "general welfare/wealth", I guess GDP will have to do. But again, like many other commonly used terms, most Americans think it implies increase in overall wealth or welfare and that isn't what it measures.
A long explanation of something so simple to understand from its name that the people who are confused don’t know how to read. GDP: Gross Domestic Product. Obviously to anyone who passed 5th grade reading comprehension it refers to what was produced domestically. If there are creatures who can’t read at the 5th grade level only a moron would take advice from them.
“Let’s talk about what GDP is. GDP is the total value of everything produced in a country:
GDP = all the stuff we produce.”
101 question, value is different from what is produced and is not a fixed number like the what is produced. 100 widgets selling for $10 each has a different value if the price goes up or down due to interest or availability in purchasing the widget. However, the number of widgets produced remains the same at 100 despite changes in value. So value as a variable can’t be equal to production which is a fixed number and not variable. What am I missing?
Noah is using “production” as shorthand for “value of production”.
In international comparisons they usually don’t want one country to count as having artificially high or low gdp because some goods are sold at a different price in the different countries, so they correct for purchasing power parity. Similarly, they try to correct for inflation across years. I think both of these measures try to differentiate change in price caused by change in quality of goods from mere change in price of the same goods but I don’t know how well they do that.
Thanks Kenny and I may be too thick or perhaps too stubborn and would be better off not pushing for clarity I may not be committed to enough to produce (create value for myself).
With your input Noah is saying the same thing twice and simply dropping the word value from the second iteration. Ostensibly, this is for clarity.
It seems I’m unclear about what value means. It seems elusive as there is value from the seller’s view and perhaps a different value from the buyer view with the ultimate value being an agreement reached between the two.
If GDP is the value of all production and value is a varying amount based on agreement, it seems GDP ends up being fundamentally the aggregate agreements about value and total production is just a multiplier of that agreement.
It makes economics seem like a swishy practice pretending to be more solid than it actually is.
I think that most economists think GDP matters mainly because it is a measurement of how much human welfare is created by all the activities going on - higher GDP means more of what people care about is going on. Figuring out how to translate market prices into what people actually care about is inherently difficult, and I think that economists who are at all philosophically inclined will agree with all of that.
But this sort of conceptual complexity is more common across all the sciences than people often imagine. Economics probably has more difficulties, because it’s trying to measure what matters to people, rather than measuring some physical thing that isn’t directly what people care about but is easier to measure. But because it is about what matters, I do think it’s still important to try.
Yes, anything can be measured if you think it can and agree on a measure. Yes, any measure ends up bound by language as it’s a product of “languaging.”
Your “but is easier to measure” recalls the story of a man on his hands and knees under a light at night in a parking lot looking for his keys. Another man joins in and after a while feeling exasperated asks the first man “Well where did you lose the keys???” The man pointed to the far unlit corner of the lot and said “I think over there but there’s very little light there.” Most anything valuable begins by being truthful.
Sorry I wasn't clear. I’m not making a claim, I’m asking a question seeking clarification. The writer is defining GDP by stating what it is. First he says it is value then he says it is equal to production. I of course do not know how is and equal are being considered by the writer or if he means to be that precise in his usage of either term when combining them in the same sentence. It’s just unclear to me at this point.
What are you claiming?
thanks so much for the clear explanation. Yet the line between what the NYT wrote, and what the BEA reports, is subtle: "... primarily reflected upward revisions to private inventory investment and federal government spending that were offset by downward revisions to nonresidential fixed investment and exports (refer to "Updates to GDP"). Imports, which are a subtraction in the calculation of GDP, were revised up." https://www.bea.gov/index.php/news/2024/gross-domestic-product-third-estimate-corporate-profits-revised-estimate-and-gdp-0
Very late to the party but isn't the purpose of minusing exports also relevant to exports too? Given the fact that you can import x, perform some value-add to x and then export that new product. Countries don't record the entire value of the export as GDP, just the value add. Otherwise you wouldn't see countries with export as a % of GDP greater than 100