"The simplest model of trade is that I am better at producing one thing, and you are better at producing something else."
Just a tiny nit that is often overlooked- trade is still beneficial even if you are actually also better at producing the thing I am personally best at producing.
The truth that you can't plug anybody into any job is why "jobs" programs tend to be a mess. During the 2008 financial crisis, a colleague said they should just put the unemployed to work building things. He reconsidered when I asked him how it would work to have a bunch of cubicle warriors laid off from Wells Fargo Finance show up on a construction site.
Some government jobs are completely sustainable, and only grow in what they consume (insert picture of Congress here). This is not free trade at work, it is the monopoly that government has over the rest of the economy. I guess if we had enough time, the parasitical relationship between government and the host economy would cause the economy to collapse, which implies that government jobs are not sustainable forever.
Overall, great article. We are indeed thinking of the labor market as a single entity and should perhaps think of it as hundreds of markets. Thanks.
Perhaps there would be less confusion if Arnold used a different term or modifier of "sustainable" to indicate the special, term of art meaning he wants to convey. Maybe "self-sustainable" would get across the idea of a patterns that "pay for themselves", so to speak, in terms of market gains from trade and thus spontaneous transaction activity that doesn't need any "help".
I think "sustainable" needs some fleshing out, but in a slightly different way.
> These patterns of trade are sustainable if and only if there are profits earned by everyone involved.
I'd suggest a positive definition of sustainable for a given period of time if it's profit maximizing for the parties involved. E.g. a pattern of trade is only sustainable if it is profit maximizing for both parties in the time-period in question.
1. I say "profit-maximizing" and not "profitable" because it's quite possible that Intel will profit whether it lays off those 15k workers or not.
2. In that respect government isn't per-se unsustainable. But if we know that trades that are wholly dependent on a temporary condition (e.g. a single, one time appropriation) aren't likely to be very sustainable.
3. In that respect, this approach fits neatly into the standard rational expectations model. If both parties operate with RE, they'll continue a pattern of trade so long as they believe it is sustainable.
4. But... it leaves a lot of room for mistake and misjudgment. People in general are rational, individuals are frequently going to guess wrong.
5. If I were to try and math this out, I'd want to devote a lot of effort to establishing the "velocity" of decision-making. I'd point out that historically, information travelled slower in innumerate ways.
Since a pattern of trade can only be sustainable for a decision-making time-frame, compressing that timeframe down necessarily makes every pattern of trade less sustainable.
"Is analysis of, or even discussion of, the economy as having one homogeneous good, one homogeneous labor input system, and one relative price (the real wage) not wrong and misleading? Is this not what leads to the false Phillips Curve analysis in which employment drives inflation?"
"When the government creates jobs in specific businesses, this is known as “industrial policy.” Industrial policy usually means having to subsidize the favored firms forever."
Well, I mean... in THEORY, there might sometimes be an argument that what the government is actually doing is paying an honest price for a non-obvious service, which the government legitimately does need to purchase at that price.
For example, I could see a very good argument that the Government might legitimately pay for, say, 20% of the costs of a major civilian shipyard to be built on american soil, so long as all the major capital equipment was certified as being rated for dual-use in both military and civilian shipyards.... With the Government then being contractually entitled to pay fair-market rates for the priority reservation of up to 5% of shipyard capacity during peacetime, and 95% during war-time.
In that instance, the Government is legitimately paying for civilian infrastructure to be built and to stay in practice NOW, because the Government might not have time to pay for it to be built LATER.
Granted, industrial policy almost never works out that smoothly or fairly in actual practice.
Economics always oversimplifies; no economic model is as complex as the actual economy. Talk about “the labor market” is an exceptionally crude oversimplification, but it is still useful in some contexts.
Wow. What set THAT off? it seems like quite the rant. A warning akin to "current thing" seems appropriate.
"In the case of labor, the price-adjustment mechanism (meaning the change in wage or salary rates) works so slowly that we usually ignore that it exists."
It's mostly a matter of framing and opinion but I would argue that S-L-O-W-L-Y is a huge factor in the labor market. It moves so slowly that the target has moved long before the market has adjusted from the target long before. But I disagree that makes it less of a market. Everyone in the market reacts to supply and demand. When prices get high, employers make substitutions. When prices get low, workers make substitutions. It is rarely if ever market clearing but actually the same is true for goods and services too. Lots of markets never clear. To take just one example, how much produce and expired products do grocery stores throw away?
As for industrial policy, I agree there are tons of mistakes. But if you say they never get it right, I say BS. The right answer is somewhere in between. Accepting there will always be mistakes, does our government do too much or too little? IDK but an example seems worthwhile. I was just recently listening to a podcast about how Taiwan got into the semi-conductor business. It was accomplished via a huge government involvement. Would it have happened without? That's a pretty certain no, at least not in Taiwan. Did the Taiwanese government subsidies pay off? Maybe not through direct returns to their treasury but I can't see an argument what they did failed in any respect.
Even to a non-economist like myself, it seems self-apparent that trying to lump all jobs into a "labor" bucket is daffy. And "Industrial policy usually means having to subsidize the favored firms forever" shows up everywhere as well - look at agricultural subsidies as a permanent fixture of U.S. economic policy, or Biden's beloved Amtrak.
At some point in high school, I noticed that the way evolutionary biology (and much of scientific reasoning) is communicated and taught to lay audiences follows the format “the organism does X to achieve Y” but in reality, the format should have been more like “if X did not achieve Y, we would not be using X as an example at all”. For many audiences and purposes, counterfactual reasoning is not very interesting or educational, offering only a narrow range of utility. So, many get an unhelpful (or limited) notion of cause and effect and learn to expect an otherwise less helpful but more acceptable view of their environment.
Arnold, note that "supply/demand" markets are mathematically feedback loops where an increase in demand creates and increase in price that signals the suppliers to increase the supply. Having systems with huge numbers of these feedback loops is standard in nature and man made systems from rockets to living cells in your body.
However, the mathematics that describes these feedback loops does show when and how they become unstable and can cause system failure. Why do I never see economists using the same mathematics to describe economies?
For example, the addition of delay times by regulators in housing markets in California (environmental impact reports) which created an instability that drives up prices/ft2 which can then oscillate and have major crashes (like 2008). Without understanding the mathematics of "control theory" economists can't understand market stability/instability or the impact of minor time changes on stability.
“Government knows how to create jobs that are not sustainable.” A little bit disingenuous because it implies that government only creates unsustainable jobs. Crossed over to polemic a little bit there.
No, this use of "sustainable" includes profitable in the sense that the value of its productivity is greater than its cost and the market demand will pay that price without any government intervention. If a job only exists because the goverment makes it exist, it is almost by definition unsustainable by the meaning of this usage of the term.
The federal and state governments had to take money from me and every other taxpayer to pay for these jobs, so, yes, they're unsustainable under Arnold's definition. Whether they might exist in an alternative profit-sensitive reality is a different question. It's very possible that they (or at least some of them) might.
And certainly all those employed by the water hustlers.
But I realize that there's a desire for the enormous sum to be made up by say, the folks who have a non-profit that offers art or music camp in the summer, and such as that. Or the local nature preserve.
Then there are the social justice groups but they punch far above their weight, culturally speaking. It's there, perhaps, that you get into the motivations of the wealthy, and their possible bitternesses.
(Going off AK's feeling about non-profits, that is, and the reliance of some on government budgets.)
Maybe there is truth to that or something like it but saintly and it's opposite don't seem quite relevant in either case. In a more important way, I think his post today shares a common trait with DEI. It has no balance. However limited, there are some good points regarding industrial policy. I don't think we want what would result from zero involvement any more than what we have now.
“However limited, there are some good points regarding industrial policy.”
Even if I accept your claim as true (and I’m not sure that I do, as public choice theory would explain), surely it matters exactly how limited the good points are.
If said good points aren’t close to 50-50, and especially if they ain’t even 20-80, then I don’t want posts that are balanced for the sake of balance.
If you prefer “balance”, ChatGPT and pretty much all LLMs are the place for you. 😏
"surely it matters exactly how limited the good points are."
Actually, it doesn't.
"If said good points aren’t close to 50-50, and especially if they ain’t even 20-80, then I don’t want posts that are balanced for the sake of balance."
I'm not entirely sure what 50-50 or 20-80 really means but it's irrelevant. It's not the reason for strongmanning industrial policy. It is the status quo. This alone earns it consideration. And “He who knows only his own side of the case knows little of that."
"The simplest model of trade is that I am better at producing one thing, and you are better at producing something else."
Just a tiny nit that is often overlooked- trade is still beneficial even if you are actually also better at producing the thing I am personally best at producing.
Ah, Ricardian "comparative advantage".
Yes, he should have inserted the word “relatively” before each instance of “better”.
The truth that you can't plug anybody into any job is why "jobs" programs tend to be a mess. During the 2008 financial crisis, a colleague said they should just put the unemployed to work building things. He reconsidered when I asked him how it would work to have a bunch of cubicle warriors laid off from Wells Fargo Finance show up on a construction site.
Some government jobs are completely sustainable, and only grow in what they consume (insert picture of Congress here). This is not free trade at work, it is the monopoly that government has over the rest of the economy. I guess if we had enough time, the parasitical relationship between government and the host economy would cause the economy to collapse, which implies that government jobs are not sustainable forever.
Overall, great article. We are indeed thinking of the labor market as a single entity and should perhaps think of it as hundreds of markets. Thanks.
Perhaps there would be less confusion if Arnold used a different term or modifier of "sustainable" to indicate the special, term of art meaning he wants to convey. Maybe "self-sustainable" would get across the idea of a patterns that "pay for themselves", so to speak, in terms of market gains from trade and thus spontaneous transaction activity that doesn't need any "help".
I think "sustainable" needs some fleshing out, but in a slightly different way.
> These patterns of trade are sustainable if and only if there are profits earned by everyone involved.
I'd suggest a positive definition of sustainable for a given period of time if it's profit maximizing for the parties involved. E.g. a pattern of trade is only sustainable if it is profit maximizing for both parties in the time-period in question.
1. I say "profit-maximizing" and not "profitable" because it's quite possible that Intel will profit whether it lays off those 15k workers or not.
2. In that respect government isn't per-se unsustainable. But if we know that trades that are wholly dependent on a temporary condition (e.g. a single, one time appropriation) aren't likely to be very sustainable.
3. In that respect, this approach fits neatly into the standard rational expectations model. If both parties operate with RE, they'll continue a pattern of trade so long as they believe it is sustainable.
4. But... it leaves a lot of room for mistake and misjudgment. People in general are rational, individuals are frequently going to guess wrong.
5. If I were to try and math this out, I'd want to devote a lot of effort to establishing the "velocity" of decision-making. I'd point out that historically, information travelled slower in innumerate ways.
Since a pattern of trade can only be sustainable for a decision-making time-frame, compressing that timeframe down necessarily makes every pattern of trade less sustainable.
And ditto for Aggregate Demand for goods and services.
As I ask in "Jackson Hole Questions"
https://thomaslhutcheson.substack.com/p/jackson-hole-questions
"Is analysis of, or even discussion of, the economy as having one homogeneous good, one homogeneous labor input system, and one relative price (the real wage) not wrong and misleading? Is this not what leads to the false Phillips Curve analysis in which employment drives inflation?"
"When the government creates jobs in specific businesses, this is known as “industrial policy.” Industrial policy usually means having to subsidize the favored firms forever."
Well, I mean... in THEORY, there might sometimes be an argument that what the government is actually doing is paying an honest price for a non-obvious service, which the government legitimately does need to purchase at that price.
For example, I could see a very good argument that the Government might legitimately pay for, say, 20% of the costs of a major civilian shipyard to be built on american soil, so long as all the major capital equipment was certified as being rated for dual-use in both military and civilian shipyards.... With the Government then being contractually entitled to pay fair-market rates for the priority reservation of up to 5% of shipyard capacity during peacetime, and 95% during war-time.
In that instance, the Government is legitimately paying for civilian infrastructure to be built and to stay in practice NOW, because the Government might not have time to pay for it to be built LATER.
Granted, industrial policy almost never works out that smoothly or fairly in actual practice.
The real estate market data is another one that makes no sense aggregating. RE markets are as varied as there are neighborhoods.
Economics always oversimplifies; no economic model is as complex as the actual economy. Talk about “the labor market” is an exceptionally crude oversimplification, but it is still useful in some contexts.
I like these posts about economic theory
Wow. What set THAT off? it seems like quite the rant. A warning akin to "current thing" seems appropriate.
"In the case of labor, the price-adjustment mechanism (meaning the change in wage or salary rates) works so slowly that we usually ignore that it exists."
It's mostly a matter of framing and opinion but I would argue that S-L-O-W-L-Y is a huge factor in the labor market. It moves so slowly that the target has moved long before the market has adjusted from the target long before. But I disagree that makes it less of a market. Everyone in the market reacts to supply and demand. When prices get high, employers make substitutions. When prices get low, workers make substitutions. It is rarely if ever market clearing but actually the same is true for goods and services too. Lots of markets never clear. To take just one example, how much produce and expired products do grocery stores throw away?
As for industrial policy, I agree there are tons of mistakes. But if you say they never get it right, I say BS. The right answer is somewhere in between. Accepting there will always be mistakes, does our government do too much or too little? IDK but an example seems worthwhile. I was just recently listening to a podcast about how Taiwan got into the semi-conductor business. It was accomplished via a huge government involvement. Would it have happened without? That's a pretty certain no, at least not in Taiwan. Did the Taiwanese government subsidies pay off? Maybe not through direct returns to their treasury but I can't see an argument what they did failed in any respect.
Even to a non-economist like myself, it seems self-apparent that trying to lump all jobs into a "labor" bucket is daffy. And "Industrial policy usually means having to subsidize the favored firms forever" shows up everywhere as well - look at agricultural subsidies as a permanent fixture of U.S. economic policy, or Biden's beloved Amtrak.
At some point in high school, I noticed that the way evolutionary biology (and much of scientific reasoning) is communicated and taught to lay audiences follows the format “the organism does X to achieve Y” but in reality, the format should have been more like “if X did not achieve Y, we would not be using X as an example at all”. For many audiences and purposes, counterfactual reasoning is not very interesting or educational, offering only a narrow range of utility. So, many get an unhelpful (or limited) notion of cause and effect and learn to expect an otherwise less helpful but more acceptable view of their environment.
Good article. Aggregation is a problem in a lot of macroeconomic concepts that can never be emphasized enough.
Excellent. You could use this as the basis for a longer essay or two in your next book.
Arnold, note that "supply/demand" markets are mathematically feedback loops where an increase in demand creates and increase in price that signals the suppliers to increase the supply. Having systems with huge numbers of these feedback loops is standard in nature and man made systems from rockets to living cells in your body.
However, the mathematics that describes these feedback loops does show when and how they become unstable and can cause system failure. Why do I never see economists using the same mathematics to describe economies?
For example, the addition of delay times by regulators in housing markets in California (environmental impact reports) which created an instability that drives up prices/ft2 which can then oscillate and have major crashes (like 2008). Without understanding the mathematics of "control theory" economists can't understand market stability/instability or the impact of minor time changes on stability.
“Government knows how to create jobs that are not sustainable.” A little bit disingenuous because it implies that government only creates unsustainable jobs. Crossed over to polemic a little bit there.
No, this use of "sustainable" includes profitable in the sense that the value of its productivity is greater than its cost and the market demand will pay that price without any government intervention. If a job only exists because the goverment makes it exist, it is almost by definition unsustainable by the meaning of this usage of the term.
So all the aerospace jobs that owe their existence to NASA were permanently unsustainable?
The federal and state governments had to take money from me and every other taxpayer to pay for these jobs, so, yes, they're unsustainable under Arnold's definition. Whether they might exist in an alternative profit-sensitive reality is a different question. It's very possible that they (or at least some of them) might.
Or perhaps all the state DOTs would be a better example.
And certainly all those employed by the water hustlers.
But I realize that there's a desire for the enormous sum to be made up by say, the folks who have a non-profit that offers art or music camp in the summer, and such as that. Or the local nature preserve.
Then there are the social justice groups but they punch far above their weight, culturally speaking. It's there, perhaps, that you get into the motivations of the wealthy, and their possible bitternesses.
(Going off AK's feeling about non-profits, that is, and the reliance of some on government budgets.)
Compared to the DEI crowd, Arnold’s rhetoric is saintly.
Maybe there is truth to that or something like it but saintly and it's opposite don't seem quite relevant in either case. In a more important way, I think his post today shares a common trait with DEI. It has no balance. However limited, there are some good points regarding industrial policy. I don't think we want what would result from zero involvement any more than what we have now.
“However limited, there are some good points regarding industrial policy.”
Even if I accept your claim as true (and I’m not sure that I do, as public choice theory would explain), surely it matters exactly how limited the good points are.
If said good points aren’t close to 50-50, and especially if they ain’t even 20-80, then I don’t want posts that are balanced for the sake of balance.
If you prefer “balance”, ChatGPT and pretty much all LLMs are the place for you. 😏
"surely it matters exactly how limited the good points are."
Actually, it doesn't.
"If said good points aren’t close to 50-50, and especially if they ain’t even 20-80, then I don’t want posts that are balanced for the sake of balance."
I'm not entirely sure what 50-50 or 20-80 really means but it's irrelevant. It's not the reason for strongmanning industrial policy. It is the status quo. This alone earns it consideration. And “He who knows only his own side of the case knows little of that."
I'll decide for myself where the place for me is.
Agreed.
As should AK re: the lack of need for “balance” when the subject is industrial policy.
You really don't see what you did, do you?
My comment was about his post. Your comment was about me. It's like if I had said he should stop writing his substack.
Thank you for this.