5 ms·

That's pretty interesting; I knew there was some criticism of SimCity's simplified tax model in the U.S. some years ago (a simple inverse relationship between t

by mjn 14y ago

That's pretty interesting; I knew there was some criticism of SimCity's simplified tax model in the U.S. some years ago (a simple inverse relationship between tax rates and business growth), but the Norwegian analysis sounds quite interesting. Any idea where I might dig up a reference to the article? I think in SimCity's case it was probably not intentional, at least originally. The first version of the game seems to have just been Will Wright coding up some ad-hoc city-simulation rules to make a fun sandbox game, not him trying to make a point about urban planning or economics on purpose. But nonetheless these interesting assumptions end up in there.

AnthonyMouse 14y ago
>I knew there was some criticism of SimCity's simplified tax model in the U.S. some years ago (a simple inverse relationship between tax rates and business growth) Isn't that pretty much accurate? I'm not aware of any serious theory postulating that raising taxes will have any non-negative effect on economic growth in and of itself. Where it breaks down is once you account for what the tax dollars buy. If government takes $5000 from you but provides $6000-equivalent worth of health services (which it may be able to do through economies of scale or reduced transaction costs or solving a collective action problem etc.) then you have a net gain of $1000 rather than a net loss of $5000. Imagine bifurcating the result: In one scenario, your country is the beneficiary of the efforts of a large international charity, you get $6000/citizen worth of health services without collecting any tax dollars from your population. Second, your country collects $5000 in taxes from each citizen and then sets the money on fire, or gifts the money to an international banking cartel in exchange for nothing whatsoever. Does anyone care to argue that the second could somehow be preferable to the first? Which brings it back to the conservative position, which is that governments through incompetence or corruption will tend to convert $5000 of tax revenue to something less than $5000 worth of services to the population, resulting in a net loss to the economy and the country's people. What so happens to be true as a factual matter is that a) governments can strike the balance in either direction depending on how well they're managed, but b) that the larger a government is, the more likely it is to allocate resources inefficiently, because as size increases economies of scale reach diminishing returns while the attention of legislators and their ability to perform due diligence is spread ever thinner over an expanding scope of government projects, leaving more opportunities for mistakes and corruption. I always thought it would be an interesting experiment to eliminate 90% of federal spending but not lower federal taxes whatsoever, just give the money to the states whose residents paid the taxes, with no strings attached (i.e. the feds serving only to enforce uniform taxation and inhibit jurisdiction shopping), and letting the states individually decide how to spend the money. You can imagine a significant reduction in corruption: California residents would no longer be paying a pro rata share of the cost of boondoggles in Massachusetts (or vice versa), and Massachusetts residents may decide they don't want them either if they have to pay the entire cost out of their own pockets.
- mjn 14y ago
  There may be some kind of relationship, but it's almost certainly not a linear one. I don't think I could summarize the research on the subject (there is a lot, and it's not my area), but it seems to strongly depend on the business, stickiness of various quantities, the specific tax rates, and other factors. Part of the problem is that economies are basically nonlinear dynamical systems, and nonlinear systems tend to have lots of strange feedback loops and attractors, where varying some inputs rarely has a simple (or often, even predictable) effect on system dynamics. An analogy might be fluid flow: high-level generalizations you might make about fluid flow are typically right only in specific flow regimes, and wrong in others (due to effects such as switching between laminar and turbulent flow). For example, if you can assume flow is always laminar, then increasing the pressure-drop across a length of pipe increases the flow rate. But increasing the pressure-drop can also induce turbulence, which may decrease the flow rate. I think most macroeconomic generalizations that assume things like "lower taxes are better for business" ignore system-dynamic effects like regime-switching. The main point, though, is that SimCity just posits a very simple relationship, rather than attempting to build a realistic model of the economy grounded in scientifically valid empirical data. This can have some propagandistic effects of basically defining a certain causal relationship as true, which may or may not be true in reality, but happens in the game's virtual world as defined. In practice, SimCity probably just did that for ease of coding and playing. An incredibly complex city simulation where your actions had unpredictable, very indirect effects, and random chance or exogenous factors or just hard-to-explain stuff overrode your policy decisions on a regular basis, might not be very fun to play, compared to an ultra-simplified world with predictable rules. But it's interesting to think of rhetorical effects as well.
  - AnthonyMouse 14y ago
    That sounds like a complicated way of explaining the broken window fallacy. "Maybe the economy will be better off if the glazier has the money because he will spend it better than the butcher whose window was broken." In specific circumstances that can really be true, but in general there is no reason to expect it to be, and unless you somehow know that it is, breaking the window is more likely to cause a net loss than a net gain. I think the "economies are complicated" explanation is the wrong one for the point you're trying to make. If you can't predict what's going to happen, and taxes usually (but not always) result in lower growth, you're usually going to lose by raising taxes. And when taxes are broad-based, this is very likely to be the case indeed. Take "usually" and average it over enough repeated iterations and you get something that looks a whole lot like "always." The place you can win just by raising taxes is where you set out to tax something that it would be economically beneficial to destroy through economic disadvantage. For example, one of the problems with sustainable energy has always been that they could achieve cost parity with fossil fuels, but only by gaining economies of scale that they couldn't do without first achieving cost parity, so you have the chicken and egg problem. If you tax fossil fuels at a sufficient rate then that in and of itself solves the chicken and egg problem, which allows sustainable energy to increase economies of scale and become even less expensive, in the long term even less expensive than untaxed fossil fuels (especially when accounting for externalities). I assume this is what you're referring to as regime-switching. But you don't, as a general rule, get that sort of thing by imposing broad-based taxes or specifically taxing anything that you wouldn't like to discourage the consumption of, you get it by correctly predicting the outcome of a targeted tax and intentionally destroying something undesirable through taxation. (Which I don't think Sim City allowed you to do.)

by mjn 14y ago

AnthonyMouse 14y ago

>I knew there was some criticism of SimCity's simplified tax model in the U.S. some years ago (a simple inverse relationship between tax rates and business growth) Isn't that pretty much accurate? I'm not aware of any serious theory postulating that raising taxes will have any non-negative effect on economic growth in and of itself. Where it breaks down is once you account for what the tax dollars buy. If government takes $5000 from you but provides $6000-equivalent worth of health services (which it may be able to do through economies of scale or reduced transaction costs or solving a collective action problem etc.) then you have a net gain of $1000 rather than a net loss of $5000. Imagine bifurcating the result: In one scenario, your country is the beneficiary of the efforts of a large international charity, you get $6000/citizen worth of health services without collecting any tax dollars from your population. Second, your country collects $5000 in taxes from each citizen and then sets the money on fire, or gifts the money to an international banking cartel in exchange for nothing whatsoever. Does anyone care to argue that the second could somehow be preferable to the first? Which brings it back to the conservative position, which is that governments through incompetence or corruption will tend to convert $5000 of tax revenue to something less than $5000 worth of services to the population, resulting in a net loss to the economy and the country's people. What so happens to be true as a factual matter is that a) governments can strike the balance in either direction depending on how well they're managed, but b) that the larger a government is, the more likely it is to allocate resources inefficiently, because as size increases economies of scale reach diminishing returns while the attention of legislators and their ability to perform due diligence is spread ever thinner over an expanding scope of government projects, leaving more opportunities for mistakes and corruption. I always thought it would be an interesting experiment to eliminate 90% of federal spending but not lower federal taxes whatsoever, just give the money to the states whose residents paid the taxes, with no strings attached (i.e. the feds serving only to enforce uniform taxation and inhibit jurisdiction shopping), and letting the states individually decide how to spend the money. You can imagine a significant reduction in corruption: California residents would no longer be paying a pro rata share of the cost of boondoggles in Massachusetts (or vice versa), and Massachusetts residents may decide they don't want them either if they have to pay the entire cost out of their own pockets.

mjn 14y ago
There may be some kind of relationship, but it's almost certainly not a linear one. I don't think I could summarize the research on the subject (there is a lot, and it's not my area), but it seems to strongly depend on the business, stickiness of various quantities, the specific tax rates, and other factors. Part of the problem is that economies are basically nonlinear dynamical systems, and nonlinear systems tend to have lots of strange feedback loops and attractors, where varying some inputs rarely has a simple (or often, even predictable) effect on system dynamics. An analogy might be fluid flow: high-level generalizations you might make about fluid flow are typically right only in specific flow regimes, and wrong in others (due to effects such as switching between laminar and turbulent flow). For example, if you can assume flow is always laminar, then increasing the pressure-drop across a length of pipe increases the flow rate. But increasing the pressure-drop can also induce turbulence, which may decrease the flow rate. I think most macroeconomic generalizations that assume things like "lower taxes are better for business" ignore system-dynamic effects like regime-switching. The main point, though, is that SimCity just posits a very simple relationship, rather than attempting to build a realistic model of the economy grounded in scientifically valid empirical data. This can have some propagandistic effects of basically defining a certain causal relationship as true, which may or may not be true in reality, but happens in the game's virtual world as defined. In practice, SimCity probably just did that for ease of coding and playing. An incredibly complex city simulation where your actions had unpredictable, very indirect effects, and random chance or exogenous factors or just hard-to-explain stuff overrode your policy decisions on a regular basis, might not be very fun to play, compared to an ultra-simplified world with predictable rules. But it's interesting to think of rhetorical effects as well.
- AnthonyMouse 14y ago
  That sounds like a complicated way of explaining the broken window fallacy. "Maybe the economy will be better off if the glazier has the money because he will spend it better than the butcher whose window was broken." In specific circumstances that can really be true, but in general there is no reason to expect it to be, and unless you somehow know that it is, breaking the window is more likely to cause a net loss than a net gain. I think the "economies are complicated" explanation is the wrong one for the point you're trying to make. If you can't predict what's going to happen, and taxes usually (but not always) result in lower growth, you're usually going to lose by raising taxes. And when taxes are broad-based, this is very likely to be the case indeed. Take "usually" and average it over enough repeated iterations and you get something that looks a whole lot like "always." The place you can win just by raising taxes is where you set out to tax something that it would be economically beneficial to destroy through economic disadvantage. For example, one of the problems with sustainable energy has always been that they could achieve cost parity with fossil fuels, but only by gaining economies of scale that they couldn't do without first achieving cost parity, so you have the chicken and egg problem. If you tax fossil fuels at a sufficient rate then that in and of itself solves the chicken and egg problem, which allows sustainable energy to increase economies of scale and become even less expensive, in the long term even less expensive than untaxed fossil fuels (especially when accounting for externalities). I assume this is what you're referring to as regime-switching. But you don't, as a general rule, get that sort of thing by imposing broad-based taxes or specifically taxing anything that you wouldn't like to discourage the consumption of, you get it by correctly predicting the outcome of a targeted tax and intentionally destroying something undesirable through taxation. (Which I don't think Sim City allowed you to do.)