3 ms·

The population curve will eventually skew, but that's not indicative of where productivity losses are happening. You don't count people that weren't born in pro

The population curve will eventually skew, but that's not indicative of where productivity losses are happening. You don't count people that weren't born in productivity, then call it a loss. You lose productivity when a real person loses the ability to produce.

thaumasiotes 9y ago
"Productivity loss" or "population decline"? Population decline from low birth rates occurs in people that weren't born. If you segment the population by age, this will be obvious -- compare the population figures to the same figures for the hypothetical situation where birth rates are higher, and you'll see that the numbers that are lower in reality than in hypotheticality are those for the young segments. Old age groups won't decline from low birth rates until all younger age groups already have. And from the perspective of losses to GDP, this is what's relevant. If birth rates decline starting in 1960, that has no effect on the population of people who were born in 1930. Their contribution to GDP will be just as it was always going to be. But the cohort born in 1980 will be much less productive, because it is smaller than it should have been, and that shows up in 2000 as a loss of young workers (compared to predictions), not old workers. > You don't count people that weren't born in productivity, then call it a loss. You do if you're arguing that losses to GDP are explained by lower birth rates.
- jdmichal 9y ago
  I agree with everything you said, except for the one axiom you state that everything else relies on: "But the cohort born in 1980 will be much less productive, because it is smaller than it should have been, and that shows up in 2000 as a loss of young workers (compared to predictions), not old workers." (Emphasis mine.) What defines what a group should have been? Predictions. But I've never heard of anyone calling a wrong prediction "losses" or "gains" before. If I'm predicted to sell a million units, and I only sell half that, did I "lose" half my sales? If a stock is predicted to go up a point, but goes down a point instead, did it "lose" two points? Comparing what materialized as reality with predictions is a perfectly fine thing to do. But differences between the two are just wrong predictions, not gains and losses. You just simply can't lose something you never had.
  - thaumasiotes 9y ago
    You've gotten hung up on the word "losses" without understanding the perspective that shows them. We're talking about a decline in GDP related to a depression in the birth rate. Imagine a society in perfect stability. Every year there are 50 people in their 20s, 48 people in their 30s, 45 people in their 40s, 41 people in their 50s, and 15 people over 60 but not yet retired. Every year, people die, causing the kind of "population losses" you're worrying about in your comments to me, but this causes no population losses or shrinkage of the kind you're talking about in your earlier comments, because those deaths are balanced out by an appropriate birth rate, keeping the population constant. Now the birth rate drops. 30 years later, there are 30 people in their 20s, 48 people in their 30s, 45 people in their 40s, 41 people in their 50s, and 15 people over 60 but not yet retired. The economy has lost 40% of the age 20-30 worker demographic (it did this over a period of ten years, but still). It has lost 0% of the age 30+ worker demographic. We asked the question "are population losses randomly distributed over the population, meaning that GDP per capita should stay constant?" and we see that the answer is no. We also asked the question "are losses concentrated among older, unproductive retirees, meaning that GDP per capita should rise?" and we see that the answer to that question is also no. Losses are concentrated among the young, which means that GDP per capita should fall when those losses age into the productive segment of the workforce. (And then rise back to exactly where it once was when those losses have aged all the way through the population, including the population of retirees.) I must say again, if we want to explain (1) a loss in GDP (the number has gotten lower) by reference to (2) a loss in the population generating the GDP (again, the number ["population"] has gotten lower), which is (3) attributed to lowered birth rates, THEN (4) the loss in population (in the sense of point 2, where we said that the population declined) has occurred among the young, not among the old. It's not even necessary to use comparisons to a hypothetical reality to perceive the losses; the bare fact that the population was once larger than it is now is a perfectly conventional interpretation of "population loss".
    - jdmichal 9y ago
      You keep attempting to avoid the fact that you must tie losses to a specific entity being lost. You lose productivity because a specific person is no longer being productive. It's as simple as that. In the case of low birth rates, you are getting productivity losses at the tail end. That's where specific productivity is being lost. Where you seem to be getting tied up is that normally that loss would be offset by gains of new productivity -- a 100% or greater population replacement rate, aka a growing population. Think of it like accounting for stock at tax time. You can't just "sell stock". You have to sell a specific stock that you bought at a specific time for a specific basis. And if you had a loss on that sale, it is immutably and forever a loss. You can, however, offset a loss by selling a different stock for a gain.

thaumasiotes 9y ago

"Productivity loss" or "population decline"? Population decline from low birth rates occurs in people that weren't born. If you segment the population by age, this will be obvious -- compare the population figures to the same figures for the hypothetical situation where birth rates are higher, and you'll see that the numbers that are lower in reality than in hypotheticality are those for the young segments. Old age groups won't decline from low birth rates until all younger age groups already have. And from the perspective of losses to GDP, this is what's relevant. If birth rates decline starting in 1960, that has no effect on the population of people who were born in 1930. Their contribution to GDP will be just as it was always going to be. But the cohort born in 1980 will be much less productive, because it is smaller than it should have been, and that shows up in 2000 as a loss of young workers (compared to predictions), not old workers. > You don't count people that weren't born in productivity, then call it a loss. You do if you're arguing that losses to GDP are explained by lower birth rates.

jdmichal 9y ago
I agree with everything you said, except for the one axiom you state that everything else relies on: "But the cohort born in 1980 will be much less productive, because it is smaller than it should have been, and that shows up in 2000 as a loss of young workers (compared to predictions), not old workers." (Emphasis mine.) What defines what a group should have been? Predictions. But I've never heard of anyone calling a wrong prediction "losses" or "gains" before. If I'm predicted to sell a million units, and I only sell half that, did I "lose" half my sales? If a stock is predicted to go up a point, but goes down a point instead, did it "lose" two points? Comparing what materialized as reality with predictions is a perfectly fine thing to do. But differences between the two are just wrong predictions, not gains and losses. You just simply can't lose something you never had.
- thaumasiotes 9y ago
  You've gotten hung up on the word "losses" without understanding the perspective that shows them. We're talking about a decline in GDP related to a depression in the birth rate. Imagine a society in perfect stability. Every year there are 50 people in their 20s, 48 people in their 30s, 45 people in their 40s, 41 people in their 50s, and 15 people over 60 but not yet retired. Every year, people die, causing the kind of "population losses" you're worrying about in your comments to me, but this causes no population losses or shrinkage of the kind you're talking about in your earlier comments, because those deaths are balanced out by an appropriate birth rate, keeping the population constant. Now the birth rate drops. 30 years later, there are 30 people in their 20s, 48 people in their 30s, 45 people in their 40s, 41 people in their 50s, and 15 people over 60 but not yet retired. The economy has lost 40% of the age 20-30 worker demographic (it did this over a period of ten years, but still). It has lost 0% of the age 30+ worker demographic. We asked the question "are population losses randomly distributed over the population, meaning that GDP per capita should stay constant?" and we see that the answer is no. We also asked the question "are losses concentrated among older, unproductive retirees, meaning that GDP per capita should rise?" and we see that the answer to that question is also no. Losses are concentrated among the young, which means that GDP per capita should fall when those losses age into the productive segment of the workforce. (And then rise back to exactly where it once was when those losses have aged all the way through the population, including the population of retirees.) I must say again, if we want to explain (1) a loss in GDP (the number has gotten lower) by reference to (2) a loss in the population generating the GDP (again, the number ["population"] has gotten lower), which is (3) attributed to lowered birth rates, THEN (4) the loss in population (in the sense of point 2, where we said that the population declined) has occurred among the young, not among the old. It's not even necessary to use comparisons to a hypothetical reality to perceive the losses; the bare fact that the population was once larger than it is now is a perfectly conventional interpretation of "population loss".
  - jdmichal 9y ago
    You keep attempting to avoid the fact that you must tie losses to a specific entity being lost. You lose productivity because a specific person is no longer being productive. It's as simple as that. In the case of low birth rates, you are getting productivity losses at the tail end. That's where specific productivity is being lost. Where you seem to be getting tied up is that normally that loss would be offset by gains of new productivity -- a 100% or greater population replacement rate, aka a growing population. Think of it like accounting for stock at tax time. You can't just "sell stock". You have to sell a specific stock that you bought at a specific time for a specific basis. And if you had a loss on that sale, it is immutably and forever a loss. You can, however, offset a loss by selling a different stock for a gain.