4 ms·

Isn't Krugman just being scientific here? ie, "We had this model for a long time, but the empirical data don't back it up anymore so it's pretty silly to stick

Isn't Krugman just being scientific here? ie, "We had this model for a long time, but the empirical data don't back it up anymore so it's pretty silly to stick with it, and we should find a better model."

carapat_virulat 11y ago
Yes, I really enjoyed the article but I am not sure what was the conclusion. Should we respect the laws of economic theory even if they have no empirical influence in the real world? Or was Krugman just acting on faith and he was finally convinced by hard data, but it took some overcoming of his bias?
cromwellian 11y ago
If only the Chicago and Austrian schools would change their minds when data supporting their propositions are weak or of low confidence. I used to be a hardcore libertarian, but all of the horrors I used to believe over marginal tax rates or regulation didn't materialize, moreover the blatant market failures and externalities both positive and negative are too hard to ignore. When you are a hard core Libertarian or Objectivist even a modest tax increase or bump in CAFE standards seems like creeping communism to be resisted at all costs. When I started looking at actual data across many countries I really couldn't see a correlation between my beliefs and the outcomes I desired. Many countries economic success or failures seem related to other factors outside the tax or regulatory regime. E.g. Silicon Valley is a bastion of innovation but also high is costs, taxes, and regulations, with a politics structure that is essentially a basket case. That's because the nexus of people and network effects and culture are more important than the tax regime unless said regime is crazy bad. Ergo, Krugman is somewhat right -- it's about people and structure (his work on international trade and industrial nexii). Too much obsession over taxes and regulation "amount" rather than on what's contextual right is what's wrong with economic debate today.
- logicchains 11y ago
  >If only the Chicago and Austrian schools would change their minds when data supporting their propositions are weak or of low confidence. Many classical and neoclassical economic propositions are of a fundamentally epistemological nature, not strictly positivist. Such as the reasoning that, if without interference person A does action X at time T, X may be taken as the action they most preferred to do at time T, as if it wasn't then why would they have done it? In which case any interference with their taking of action X necessarily produces an inferior outcome in terms of their preferences at time T, as they're forced to take an action other than their most-preferred action. In this sense a person's preference for an action is defined by their taking of that action over all other possible ones when not interfered with, and as a definition its validity is a matter of epistemology, not something subject to empirical analysis. The Austrian school uses this ordinal form of utility as its basis, building upon the idea that interfering with someone's free action is the one thing that can clearly be shown to be undesirable as above. The Chicago school, on the other hand, also incorporates a quantitative form of utility, to allow for interpersonal utility comparisons. This requires some epistemological assumptions, for instance that preferences don't change over time or their rate of change follows a particular mathematical function, or that if someone buys 3 apples for $30 they value each at $10, rather than e.g. the first at $13, the second at $2, and the third at $15. Austrians dispute the validity of any such assumptions, and hence reach different conclusions. These epistemological foundations are matters of logic/philosophy, and hence no more amenable to empirical analysis than a proposition like 'abortion is immoral'. Modern economics tends to ignore epistemological issues, and instead focuses on finding the best way to maximise particular empirical metrics, without going into detailed study of the epistemological and philosophical validity of the metrics in question. So it's important to distinguish between economics as a tool for predicting and assessing the effects of particular phenomena on particular metrics (econometrics) and economics as an epistemological approach.
  - DanielBMarkham 11y ago
    This is a hugely important point, and couched as it is in somewhat difficult terminology it may not reach the audience it should. Polemicists have the job of explaining meaning to others while persuading them. I will show you various pieces of data supporting my thesis and my thesis will give you understanding of how and why things work. If you write an opinion column, this is your job. This is not the same at all as trying to minimize assumptions and simply show relationships.
  - bcg1 11y ago
    Most thoughtful, intelligent, and logical response I've seen on an economics thread on this forum. Thank you
  - cromwellian 11y ago
    >Such as the reasoning that, if without interference person A does action X at time T, X may be taken as the action they most preferred to do at time T, as if it wasn't then why would they have done it? I'm well aware of the differences between the Chicago and Austrian schools, but I think this quote is a dubious assumption because it assumes human beings are primarily rational creates that act according to rules and reason. It's actually possible for people to take actions which they don't prefer to take. Praxeology IMHO is thoroughly unscientific, contradicts modern neuroscience and psychology with many of its assumptions about people, and a dubious exercise, and its use by Austrians to argue against public policy actions extremely weak. To me the purpose of economics is to study the real world as close as you can, and make predictions that can be used to not only understand our condition or future problems, but to make adjustments so as to create better outcomes. It is not to provide an ideological, philosophical basis for those actions, or for no actions at all. It's the difference between Medicine and Bioethics. Both the Chicago school and the Austrian school start out with assuming the outcomes -- a free market s what's desired, and produced the best outcomes. They they work backwards using different methods to justify their conclusions. I prefer not making the assumption that the unfettered free market produces the best outcome always. For example, when externalities are not priced in for things human beings care about in the abstract but not in their day to day preferences, like the long term ecosystem, or the health and welfare of future generations, I prefer a system of economics that works on a case by case basis, looking at problems, and figuring out experiments to run or course corrections. If economics is unpredictable, or the consequences of actions unknowable it also means the consequences of inaction or the 'free market' are unknowable, but that's no basis for siding against government action. We live in a complex evolutionary system, and it may be the only way to get out of local optima is to run lots of experiments that fail, including government intervention. That's why I think a mixed economy is a more evolutionary stable strategy than a pure free market or marxist one. I also think it's more moral than either of those.
- aiscott 11y ago
  The effects of marginal tax rates materialized in the 70's and early 80's. It was often smarter to have just one person work in a family because the take home from a second job would be entirely absorbed in taxes and the costs of working (car, gas, etc).
notahacker 11y ago
Some of it's that and some of it's the fact that he's addressing different arguments, and from somewhat shifted political perspectives Both Krugman's quoted passages acknowledge that conventional economic theory would tend to suggest that minimum wages reduce the size of the work place, but both labour economists' theories of labour supply and some empirical work suggests that this isn't universally the case, especially where increases are small. Neither say very much on the subject. But the 1998 one is attacking a book which (according to Krugman) suggests that compelling employers to pay more - 57% more in the example - will reduce welfare costs and have no adverse effect on employment The recent article attacks a more general view that wage reduction is essential to get the nation's economy out of a rut, and is particularly approving of quasi market-based mechanisms to raise minimum wages (trade unions and Walmart's justification in terms of efficiency wages) It's the political line of argument rather than the use of economic theory that has changed (and given that Krugman doesn't exactly pull his punches, it's not surprising that one article reads almost like an attack on the other). He's not having to turn his back on economic theory to argue that the US would be better off if minimum wages were a little higher, and it wasn't slavish devotion to economic theory that caused him to contest what he described as a "moral" argument for wage regulation over redistribution from Pollin and Lice.
bsbechtel 11y ago
If you talk to any economist that does research for a living, they all will acknowledge that the data used in empirical studies in the field of economics is notoriously unreliable. Chief among the reasons for this is that it is impossible to collect data on all the different factors that go into a human's decision making process, and also that there is an observer effect (similar to that in physics) that influences data collection in many types of economic studies, especially forward facing ones.
- bandrami 11y ago
  they all will acknowledge that the data used in empirical studies in the field of economics is notoriously unreliable Then economics absolutely needs to stop being called a "science".
  - bsbechtel 11y ago
    Data in other fields is unreliable as well, just less so, and we are constantly coming up with new technologies to improve our capabilities for data collection in the hard sciences. Data collection is just one aspect of science. Should we eliminate the field of theoretical physics because they can't collect data and run experiments to test their theories?
xixi77 11y ago
the point of the article is that this is what the author thought was the case, but not what actually appears to be the case -- e.g. it seems that Krugman's interpretation of the same Card & Krueger evidence has changed, as opposed to him changing his view based on new empirical evidence available since then (which is quite messy and inconclusive, tbh).
- bandrami 11y ago
  In physics, when a new study is released physicists don't change their opinions overnight. There's a big difference in how a scientist treats a study that was recently released and one that has survived a decade or so.

carapat_virulat 11y ago

Yes, I really enjoyed the article but I am not sure what was the conclusion. Should we respect the laws of economic theory even if they have no empirical influence in the real world? Or was Krugman just acting on faith and he was finally convinced by hard data, but it took some overcoming of his bias?

cromwellian 11y ago

If only the Chicago and Austrian schools would change their minds when data supporting their propositions are weak or of low confidence. I used to be a hardcore libertarian, but all of the horrors I used to believe over marginal tax rates or regulation didn't materialize, moreover the blatant market failures and externalities both positive and negative are too hard to ignore. When you are a hard core Libertarian or Objectivist even a modest tax increase or bump in CAFE standards seems like creeping communism to be resisted at all costs. When I started looking at actual data across many countries I really couldn't see a correlation between my beliefs and the outcomes I desired. Many countries economic success or failures seem related to other factors outside the tax or regulatory regime. E.g. Silicon Valley is a bastion of innovation but also high is costs, taxes, and regulations, with a politics structure that is essentially a basket case. That's because the nexus of people and network effects and culture are more important than the tax regime unless said regime is crazy bad. Ergo, Krugman is somewhat right -- it's about people and structure (his work on international trade and industrial nexii). Too much obsession over taxes and regulation "amount" rather than on what's contextual right is what's wrong with economic debate today.

logicchains 11y ago
>If only the Chicago and Austrian schools would change their minds when data supporting their propositions are weak or of low confidence. Many classical and neoclassical economic propositions are of a fundamentally epistemological nature, not strictly positivist. Such as the reasoning that, if without interference person A does action X at time T, X may be taken as the action they most preferred to do at time T, as if it wasn't then why would they have done it? In which case any interference with their taking of action X necessarily produces an inferior outcome in terms of their preferences at time T, as they're forced to take an action other than their most-preferred action. In this sense a person's preference for an action is defined by their taking of that action over all other possible ones when not interfered with, and as a definition its validity is a matter of epistemology, not something subject to empirical analysis. The Austrian school uses this ordinal form of utility as its basis, building upon the idea that interfering with someone's free action is the one thing that can clearly be shown to be undesirable as above. The Chicago school, on the other hand, also incorporates a quantitative form of utility, to allow for interpersonal utility comparisons. This requires some epistemological assumptions, for instance that preferences don't change over time or their rate of change follows a particular mathematical function, or that if someone buys 3 apples for $30 they value each at $10, rather than e.g. the first at $13, the second at $2, and the third at $15. Austrians dispute the validity of any such assumptions, and hence reach different conclusions. These epistemological foundations are matters of logic/philosophy, and hence no more amenable to empirical analysis than a proposition like 'abortion is immoral'. Modern economics tends to ignore epistemological issues, and instead focuses on finding the best way to maximise particular empirical metrics, without going into detailed study of the epistemological and philosophical validity of the metrics in question. So it's important to distinguish between economics as a tool for predicting and assessing the effects of particular phenomena on particular metrics (econometrics) and economics as an epistemological approach.
- DanielBMarkham 11y ago
  This is a hugely important point, and couched as it is in somewhat difficult terminology it may not reach the audience it should. Polemicists have the job of explaining meaning to others while persuading them. I will show you various pieces of data supporting my thesis and my thesis will give you understanding of how and why things work. If you write an opinion column, this is your job. This is not the same at all as trying to minimize assumptions and simply show relationships.
- bcg1 11y ago
  Most thoughtful, intelligent, and logical response I've seen on an economics thread on this forum. Thank you
- cromwellian 11y ago
  >Such as the reasoning that, if without interference person A does action X at time T, X may be taken as the action they most preferred to do at time T, as if it wasn't then why would they have done it? I'm well aware of the differences between the Chicago and Austrian schools, but I think this quote is a dubious assumption because it assumes human beings are primarily rational creates that act according to rules and reason. It's actually possible for people to take actions which they don't prefer to take. Praxeology IMHO is thoroughly unscientific, contradicts modern neuroscience and psychology with many of its assumptions about people, and a dubious exercise, and its use by Austrians to argue against public policy actions extremely weak. To me the purpose of economics is to study the real world as close as you can, and make predictions that can be used to not only understand our condition or future problems, but to make adjustments so as to create better outcomes. It is not to provide an ideological, philosophical basis for those actions, or for no actions at all. It's the difference between Medicine and Bioethics. Both the Chicago school and the Austrian school start out with assuming the outcomes -- a free market s what's desired, and produced the best outcomes. They they work backwards using different methods to justify their conclusions. I prefer not making the assumption that the unfettered free market produces the best outcome always. For example, when externalities are not priced in for things human beings care about in the abstract but not in their day to day preferences, like the long term ecosystem, or the health and welfare of future generations, I prefer a system of economics that works on a case by case basis, looking at problems, and figuring out experiments to run or course corrections. If economics is unpredictable, or the consequences of actions unknowable it also means the consequences of inaction or the 'free market' are unknowable, but that's no basis for siding against government action. We live in a complex evolutionary system, and it may be the only way to get out of local optima is to run lots of experiments that fail, including government intervention. That's why I think a mixed economy is a more evolutionary stable strategy than a pure free market or marxist one. I also think it's more moral than either of those.
aiscott 11y ago
The effects of marginal tax rates materialized in the 70's and early 80's. It was often smarter to have just one person work in a family because the take home from a second job would be entirely absorbed in taxes and the costs of working (car, gas, etc).

notahacker 11y ago

Some of it's that and some of it's the fact that he's addressing different arguments, and from somewhat shifted political perspectives Both Krugman's quoted passages acknowledge that conventional economic theory would tend to suggest that minimum wages reduce the size of the work place, but both labour economists' theories of labour supply and some empirical work suggests that this isn't universally the case, especially where increases are small. Neither say very much on the subject. But the 1998 one is attacking a book which (according to Krugman) suggests that compelling employers to pay more - 57% more in the example - will reduce welfare costs and have no adverse effect on employment The recent article attacks a more general view that wage reduction is essential to get the nation's economy out of a rut, and is particularly approving of quasi market-based mechanisms to raise minimum wages (trade unions and Walmart's justification in terms of efficiency wages) It's the political line of argument rather than the use of economic theory that has changed (and given that Krugman doesn't exactly pull his punches, it's not surprising that one article reads almost like an attack on the other). He's not having to turn his back on economic theory to argue that the US would be better off if minimum wages were a little higher, and it wasn't slavish devotion to economic theory that caused him to contest what he described as a "moral" argument for wage regulation over redistribution from Pollin and Lice.

bsbechtel 11y ago

If you talk to any economist that does research for a living, they all will acknowledge that the data used in empirical studies in the field of economics is notoriously unreliable. Chief among the reasons for this is that it is impossible to collect data on all the different factors that go into a human's decision making process, and also that there is an observer effect (similar to that in physics) that influences data collection in many types of economic studies, especially forward facing ones.

bandrami 11y ago
they all will acknowledge that the data used in empirical studies in the field of economics is notoriously unreliable Then economics absolutely needs to stop being called a "science".
- bsbechtel 11y ago
  Data in other fields is unreliable as well, just less so, and we are constantly coming up with new technologies to improve our capabilities for data collection in the hard sciences. Data collection is just one aspect of science. Should we eliminate the field of theoretical physics because they can't collect data and run experiments to test their theories?

xixi77 11y ago

the point of the article is that this is what the author thought was the case, but not what actually appears to be the case -- e.g. it seems that Krugman's interpretation of the same Card & Krueger evidence has changed, as opposed to him changing his view based on new empirical evidence available since then (which is quite messy and inconclusive, tbh).

bandrami 11y ago
In physics, when a new study is released physicists don't change their opinions overnight. There's a big difference in how a scientist treats a study that was recently released and one that has survived a decade or so.