Another tax strawman.
(Cross-posted at the Galloping Beaver.)
John McIntyre, railing against those who favour higher US taxes, says:
Why have the U.S. and Britain shown tremendous growth since Reagan and Thatcher while Continental Europe with their pseudo-socialism putters along with chronic double-digit unemployment? Go around the world these last 25 years and compare nations with high tax rates to countries with low tax rates, you'll find a pattern.
Absolutely you will. There is such thing as the equity-efficiency tradeoff, and it's at the heart of the democratic economic decision. I believe that Sweden pays for the high rate of social insurance it provides its citizens with some measure of long-run growth and economic dynamism. Making that tradeoff between growth and publicly guaranteed economic equality is something that majorities in Sweden have traditionally accepted and that majorities of Americans have rejected.
To further underscore McIntyre's point, from the Fraser Institute's latest TaxFacts, here are the major Western countries ranked by the share of their GDP that goes to taxes. With the possible exception of Mexico, the countries at the top of the list tend to be the fastest-growing, and those at the bottom are the most redistributive:
Mexico: 19%
Korea 25.3%
Japan 25.3%
United States: 25.6%
Switzerland: 29.5%
Ireland: 29.7%
Slovak Republic: 31.1%
Australia: 31.6%
Turkey: 32.8%
Canada: 33.8%
Poland: 34.2%
Spain: 34.9%
New Zealand: 34.9%
Germany: 35.5%
United Kingdom: 35.6%
Greece: 35.7%
Portugal: 37.1%
Czech Republic: 37.7%
Hungary: 38.5%
Netherlands: 38.8%
Iceland: 39.8%
Luxembourg: 41.3%
Austria: 43.1%
Italy: 43.1%
France: 43.4%
Norway: 43.4%
Finland: 44.8%
Belgium: 45.4%
Denmark: 48.3%
Sweden: 50.6%
The list also shows the big flaw in McIntyre's argument: the differences in growth and economic insurance take place over a massive range in national tax shares -- 30% of a country's GDP. But, for better or worse, very few North American liberals are actually advocating the kind of economic upheaval that would reroute 50% of GDP through the government. US liberals don't argue for the US to adopt overall tax rates comparable to those found in Sweden or even France. McIntyre's traget, Sabastian Mallaby, certainly doesn't. Few of the bloggers or pundits I read have Denmark in mind when they complain about tax giveaways to corporations or to the rich or to the amazing shelf-life of long-discredited supply side nostrums.
A more relevant way to approach the issue is to examine the sort of highly developed, low-union-density, limited-regulation, free-market economies that are actually comparable to the US... like Canada. From the list above, Canada and the UK (since McIntyre brought it up) have, respectively, gross tax rates of 8.2% and 10% higher than the current gross US rate. At least as between Canada and the US, the ratio has been fairly stable since 1994, though federal rates at least have fallen since then in both countries. From the US Department of Labor's statistics (Table 1), the US GDP per capita grew 24.0% between 1994 and 2004. Canada's grew by 25.9%. The UK's grew by 28.0%.
Raise the US tax take to Sweden's 51% share and I'm in full agreement that, in the long run, Americans will pay a massive price in terms of GDP growth and entrepreneurial dynamism. It's not something I would vote for. Raise the tax share to Canada's modest 34% or the UK's 35% -- rates that could wipe out the budget deficit and cover a pretty nice universal public health care plan to boot -- and the international evidence that the US would pay much of a price in terms of long-run growth gets a lot weaker. That's a risk I'd be willing to take.
Update: For the record, a commenter points out that Sweden's growth isn't too shabby -- and using the numbers from the same BoL table, Sweden has actually outgrown the US in % terms since 1994. So even if nobody is arguing for turning the US into Sweden, it's still not the best country to pick on right now. There's always France...
John McIntyre, railing against those who favour higher US taxes, says:
Why have the U.S. and Britain shown tremendous growth since Reagan and Thatcher while Continental Europe with their pseudo-socialism putters along with chronic double-digit unemployment? Go around the world these last 25 years and compare nations with high tax rates to countries with low tax rates, you'll find a pattern.
Absolutely you will. There is such thing as the equity-efficiency tradeoff, and it's at the heart of the democratic economic decision. I believe that Sweden pays for the high rate of social insurance it provides its citizens with some measure of long-run growth and economic dynamism. Making that tradeoff between growth and publicly guaranteed economic equality is something that majorities in Sweden have traditionally accepted and that majorities of Americans have rejected.
To further underscore McIntyre's point, from the Fraser Institute's latest TaxFacts, here are the major Western countries ranked by the share of their GDP that goes to taxes. With the possible exception of Mexico, the countries at the top of the list tend to be the fastest-growing, and those at the bottom are the most redistributive:
Mexico: 19%
Korea 25.3%
Japan 25.3%
United States: 25.6%
Switzerland: 29.5%
Ireland: 29.7%
Slovak Republic: 31.1%
Australia: 31.6%
Turkey: 32.8%
Canada: 33.8%
Poland: 34.2%
Spain: 34.9%
New Zealand: 34.9%
Germany: 35.5%
United Kingdom: 35.6%
Greece: 35.7%
Portugal: 37.1%
Czech Republic: 37.7%
Hungary: 38.5%
Netherlands: 38.8%
Iceland: 39.8%
Luxembourg: 41.3%
Austria: 43.1%
Italy: 43.1%
France: 43.4%
Norway: 43.4%
Finland: 44.8%
Belgium: 45.4%
Denmark: 48.3%
Sweden: 50.6%
The list also shows the big flaw in McIntyre's argument: the differences in growth and economic insurance take place over a massive range in national tax shares -- 30% of a country's GDP. But, for better or worse, very few North American liberals are actually advocating the kind of economic upheaval that would reroute 50% of GDP through the government. US liberals don't argue for the US to adopt overall tax rates comparable to those found in Sweden or even France. McIntyre's traget, Sabastian Mallaby, certainly doesn't. Few of the bloggers or pundits I read have Denmark in mind when they complain about tax giveaways to corporations or to the rich or to the amazing shelf-life of long-discredited supply side nostrums.
A more relevant way to approach the issue is to examine the sort of highly developed, low-union-density, limited-regulation, free-market economies that are actually comparable to the US... like Canada. From the list above, Canada and the UK (since McIntyre brought it up) have, respectively, gross tax rates of 8.2% and 10% higher than the current gross US rate. At least as between Canada and the US, the ratio has been fairly stable since 1994, though federal rates at least have fallen since then in both countries. From the US Department of Labor's statistics (Table 1), the US GDP per capita grew 24.0% between 1994 and 2004. Canada's grew by 25.9%. The UK's grew by 28.0%.
Raise the US tax take to Sweden's 51% share and I'm in full agreement that, in the long run, Americans will pay a massive price in terms of GDP growth and entrepreneurial dynamism. It's not something I would vote for. Raise the tax share to Canada's modest 34% or the UK's 35% -- rates that could wipe out the budget deficit and cover a pretty nice universal public health care plan to boot -- and the international evidence that the US would pay much of a price in terms of long-run growth gets a lot weaker. That's a risk I'd be willing to take.
Update: For the record, a commenter points out that Sweden's growth isn't too shabby -- and using the numbers from the same BoL table, Sweden has actually outgrown the US in % terms since 1994. So even if nobody is arguing for turning the US into Sweden, it's still not the best country to pick on right now. There's always France...
9 Comments:
Politics is about the distribution of goods, just as economics is. All government is redistributive. In the U.S., the redistribution is entirely in the interest of the corporate executive class, the class, which finances and controls the Republican Party.
If social insurance was such a sure damper on economic growth, Sweden would have been reduced to poverty years ago. The fact is, Sweden enjoys pretty good economic growth, for the most part. The U.S. has had excellent economic growth, with the small niggly little detail that almost all the benefits from growth have flowed to the top 1/2 of 1%.
You can have low taxes, and dampen economic dynamism. Repealing the estate tax, as the Republicans advocate, will almost certainly dampen economic mobility and dynamism. Reagan's combination of tax cuts for the Rich and record tax increases on the working class resulted in fairly slow growth.
No; not all government is redistributive. It's only redistributive if it takes income from the people who own the factors of production (labour, land, capital) and give it to those who don't -- through taxation. You can also redistribute to the rich through various corporate subsedies, but most millionaire tax breaks are better thought of as ways for the rich to avoid paying for redistribution, not ways of redistributing to the rich. When the rich can shelter income in roth iras, they just keep more of what they make in the legal property rights regime that exists.
Also, you don't have to see countries "reduced to poverty" for McIntyre's argument to work. High-tax countries just have to grow slower than countries with lower tax rates and less corresponding income redistribution. From my understanding, that's largely been the case, and you can certainly overtax an economy. You're right that Sweden is both pretty rich and has been growing fine since 1994 -- France would have been a better example. It's still perfectly credible that they pay for their equality with growth -- and that's a political decision to be made by voters. At any rate, with such a huge public sector they're a very different sort of economy.
Finally, I certainly agree that taxing the rich is less hurtful, through its demand effects, than taxing the poor. But even in the US, the rich pay the lion's share of actual taxes. That's as it has to and should be. Arguments about aggregate amounts of taxation are therefore mainly about how much tax falls to those who can pay it: the rich. The US could currently raise its taxes on the rich a lot without risking a slowdown -- US taxes are so low by global standards there's just a lot of room for more tax. Could they raise them to Swedish levels without bringing on stagnation? Quite possibly not, but very few people are arguing for that.
Another interesting point of view on this is that the USA statistics are an average of the statististics of a fairly prosperous fairly ''canadian'' nation and a fairly backward ''central american'' (black and brown skinned) one, and that this skews voting patterns too:
http://neweconomist.blogs.com/new_economist/2006/05/john_roemer.html
"It's only redistributive if it takes income from the people who own the factors of production (labour, land, capital) and give it to those who don't"
Look at what you are saying here: Government is "only" redistributive if it benefits the slaves?
Do you seriously think the details of "the legal property rights regime" have no distributive implications? Bush is constantly tinkering with tort law "reform" and changes in labor law affecting payment for overtime, not to mention Social Security "reform" and corporate pension guarantees and student loans, and mine safety and OSHA and logging the national forests and and and. What do you think all of that is about, if not redistribution of income?
Seriously, your worldview needs a bit of re-examination. Just as a matter of pure economic theory, risk and the availability of insurance matters a lot to income distribution, and government, for a variety of reasons, is the best or only provider of a great variety of forms of insurance, not to mention other risk-dampeners, like police protection and public education.
France has made a different choice from the United States, with regard to the amount of leisure its population will enjoy. For a variety of reasons, in a highly organized (and, therefore, highly structured) economy, that becomes a collective choice for most people. The average Frenchman has more vacation time and a smaller home with less stuff in it than the average American, who works a lot more. The difference in time spent working accounts for most of the difference in median incomes.
Logically, any claim about (static? cross-sectional?) differences in standard of living should be distinguishable in character from a claim about (dynamic? time-series?) growth rates.
Presumably, there's some rate of technological/scientific progress, which imposes a ceiling or frontier on potential national (per capita) income. For an underdeveloped country that ceiling is far away and unimportant, compared to the ability to form and accumulate capital and to reform institutions. But, for developed countries, that ceiling (or frontier) must be fairly close. My guess is that, while growth rates will vary, and different countries are aiming at different equilibrium levels of per capita income/leisure, most developed countries will tend to keep fairly close to the frontier of technological progress.
You seem to be claiming something else, something which would require that more "socialist" (high-tax) countries gradually fall further and further away from the frontier of technological progress, because their higher provision of social insurance (with its redistribution effects) somehow makes them unable to assimilate technological progress.
The devil is in the details, of course. I can accept the idea that one could devise a set of institutions, which would provide a lot of social insurance and also impede assimilation of technological progress. I would also suggest that one could devise a set of authoritarian institutions, which would direct almost all income to a small elite, and which would impede assimilation of technological progress. But, that's not your model, or your evidence.
Sweden had cradle-to-grave socialism for most of the 20th century, and somehow its economic development and growth has managed to keep pace generally with technological progress.
I don't think it is such a stretch to imagine that the ability to assimilate technological progress into economic growth thru capital accumulation might actually be enhanced by higher rates of taxation and better provision of social insurance (i.e. better public education, subsidies for university education, health care, better infrastructure) than is current practice in the U.S.
I am sorry, but I am not much impressed by your "what the rich get is theirs "naturally", and what everyone else gets, is negotiable" attitude.
He, who makes the rules, gets the gold, so he, who has the gold, has probably been making the rules.
There may be an equity-efficiency tradeoff, but, barring extreme cases (like Mexico - which has traded off so much equity that it has become inefficient (corrupt and uneducated) as well) that tradeoff exists at the level of specfic policies, not at a macro-level.
For example, the U.S. could easily significantly increase both its equity and efficiency by adopting a single payer national health system.
Funding its elementary school system at a higher level of government could probably accomplish the same thing.
In Canada, a national pharmacare program or a national securities regulator would likely increase both equity and efficiency as well.
On the other hand, poorly designed government programs (mandatory sentencing laws, preferring military solutions to diplomatic ones, regional pork barrel programs (bridges to nowhere) etc.) can decrease both equity and efficiency.
Keeping this in mind, it is not surprising that, as far as I know, studies have found little correlation between size of government sector and long run growth, or to put it another way Go around the world these last 25 years and compare nations with high tax rates to countries with low tax rates, you won't find any pattern.
Your chart supports this as well, with many countries at the top like Mexico, Japan and Turkey that are poor or struggling and countries at the bottom like Holland, Norway, Finland and Denmark that are wealthy and growing rapidly.
Looking at the BoL document you link to (table 2), it seems that, starting from 1960, the only countries which have lost ground against the U.S. are Australia and England. The Japanese and Koreans (who were way behind to start with) have made significant gains, while the Norwegians (oil), the French, Austrians and Italians (postwar rebuilding?) have made decent gains as well. If there is any pattern here, it is one of general convergence, regardless of size of government - which likely goes to Bruce's comment about a 'technological frontier'.
---
One also has to keep in mind that because government provided services are often transacted far far below fair market value (compare the cost of maintaining a road network to a Ministry of Transportation vs. the revenue that could be generated if every road was tolled), there is a bias in that reported GDP will be higher relative to total economic welfare in countries with smaller government sectors than in those with larger government sectors (assuming that the larger sector consists of services provided below market cost).
In the same vein, countries with stronger social networks where more transactions occur outside of the market will have lower recorded GDP as well. One could argue that a society where people can't rely on themselves or friends and family to take care their old and young relatives is a society which is breaking down, but it is also a society where GDP goes up as people pay babysitters and put people in old age homes.
Plus there is the usual concern about purchasing power parity adjustments and so on (for example, in table 2, Canada's GDP vs. the U.S. GDP tracks the exchange rate quite closely, despite the chart being adjusted to reflect purchasing power parity.)
One also has to consider whether it is more 'efficient' to have somebody be poor, indebted or scared enough to work 50 hours a week and not take any vacations in one country vs. the same person having a comfortable life and choosing to work 35 hours a week and take 6 weeks of vacation in another country.
It is difficult to distinguish at a macro level between someone who chooses not to work because they can afford to make that decision and someone who needs to work but can't find a job.
Also, as Bruce is saying, there are a lot of government interventions which don't show up as taxes. For example, the U.S. has pretty draconian intellectual property monopoly laws. Again this drives up GDP while reducing actual prosperity by forcing people to pay monopoly prices for more things. If you were to add up all the corporate profits earned in the U.S. from government protected monopolies (TV, Film, Pharamceuticals, Software, etc.) it would make up a pretty big chunk of the economy. One might question whether the government intervention which allowed Bill Gates to become the richest person in the world really served either equity or efficiency.
---
It is also worth considering other factors which influence government size around the world. When I wrote a post on this a while back, I found UN studies which suggest that, with regards to size of government operations, countries with small geographic areas and large populations tended to have smaller government sectors. Small geographic areas likely leads to lower infrastructure costs, larger population leads to economies of scale.
With respect to transfers (e.g. welfare, old age pensions etc.) it is typically governments which have more open economies (a high trade vs gdp ratio) which put more into transfers, perhaps to cushion their citizens from the impact of competing globally.
Finally, as an aside, you commented that, "But even in the US, the rich pay the lion's share of actual taxes."
There is an interesting chart here which suggests that the amount of prgressivity in the U.S. tax system is quite a bit less than in most other developed nations. It also notes that pre-tax, the level of inequality in Sweden and the U.S. is not very different. I guess Canada's tax system, already one of the less progressive ones, will get less so as long as the Conservatives are in power.
---
P.S. I think Bruce's comment about all government being redistributive was just meant to reflect that all government action creates winners and losers. Even putting in a traffic light will help people who drive on the less busy street at the expense of the people who drive on the busy street (even if it helps both sets of drivers, it will help one set more, and hence redistrubiute relative well-being).
With regard to Bruce's point about government intervention in the U.S. benefitting the wealthy, this new (and free online) book by economist Dean Baker is worth a read.
Thanks for the comments, all of you. Although there's just too much in both the last two posts to respond to.
Bruce -- you're simply reading too much into this post. It's about aggregate tax and the political arguments surrounding the optimal tax rate -- not the structure of taxes, not the debate over liberal vs. corporatist vs. socialist institutions. In liberal economies, for a given property rights regime, redistribution is carried out mainly through tax and transfer, in cash or in kind. Politics also has a role in tinkering with the property rights regime (particularly I think by encouraging a more diverse set of people to hold capital), but that's not what I'm writing about at all. If I own property in Canada, I take the income from it. My income is then subject to taxation. I personally think we could make the economies in Canada and especially the US a whole lot fairer without fundamentally altering the prevailing property-rights framework, through taxation (and other types of redistribution like encouraging a more diverse set of people to hold capital). Despite what McIntyre argues, there's plenty of room to do that without hurting growth; similar liberal economies have far higher aggregate tax rates for similar levels of growth. And it doesn't require adopting the institutions of Sweden or France, or raising the tax revenue required by those institutions. Whether we want to turn the US or Canada into Sweden or France is a different question. But I'd wager that most North American liberals don't.
Declan -- a couple of things. First, thanks for the link to the chart. But it sort of proves my point: given a basic capitalist private-property setup the tax system by in large determines the level of equality through redistribution. Regarding progressivity, the US system is wildly unprogressive and even regressive for parts of the income distribution (last time I checked anyway.) But even without progressivity, the rich by definition pay the most tax, and tax hikes and cuts are almost always geared toward the rich in basic revenue terms (Reagan's payroll tax hike being an exception. The more unequal the society, the more this is the case. Even "good" tax cuts like the 1986 US one mainly benefit the rich.) I'm not saying anything here about the structure of the tax system -- just that debates over what the aggregate level of taxation is are mainly arguments over how much the rich get taxed.
Also, I looked at the numbers from 1960 and thought about writing about growth rates vs. overall wealth and taxes; the problem is that overall wealth is so much a product of history that current tax policy can't tell much I don't think. Actually, I crunched the 2000-2004 numbers and found a pretty good (negative) correlation between the current tax policy and recent growth rates for the whole range of countries. Korea and Japan and very recently Mexico are all growing at a good clip under the low-tax regimes.
For the more general question of whether high taxes slow down an economy for a given institutional framework, I'm interested that you've read the answer is basically no. My understanding has been that the answer is a qualified yes (of course it's possible to have bad low-tax policy or to misdirect the taxes). Now it seems to me that the richer we get, the more weight should be assigned to the "equity" over the "efficiency", and that economics doesn't have a good formal way to deal with that idea. Growth is certainly not my field of expertise, but I do believe in the fundamental tradeoff -- that taxes distort economic decisions. (I would!) I'd be really interested in economic literature that directly disputes it.
Oh yeah, Blissex, good point and thanks for the link. That's a terrific blog. I hadn't seen it before.
Sorry, I don't have any good links off-hand, but what I have seen has been mostly empirical evidence looking at growth rates across countries over time and within countries where tax takes have varied and finding no significant correlation between tax rate and growth rate within the developed world.
At a basic level, we know that 0% and 100% taxation are both pretty ineffective (especially 0%!), so the optimum is somewhere in between. Where that optimal point is for any given country depends on the specific policies that increased/decreased taxes would fund/not fund as well as country specific factors like what I mentioned in my previous post.
I don't see any reason to think that the U.S. is not in a position where the growth rate could be increased with higher tax rates (assuming smart policies, not using the money to invade middle eastern countries and build a wall on the Mexican border.)
It is true that taxes can distort economic decisions, but so what?
Sure it may be plausible that millionaires would stop trying to pile up cash if the highest bracket was a couple of points higher, but who is to say that the economic benefit from some person getting the medical treatment they need from the taxes taken from that rich person don't offset the loss because that rich person stopped working so hard because of the high tax rate. Or that the road built with the increased tax dollars doesn't facilitate a greater expansion than the losses from reduced incentives.
And of course, I haven't seen much empirical evidence that tax rates influence people the way economists seem to assume they do.
I speak only for myself here, but I know that if I had enough money to retire, I would. Lower taxes will help me get to that point of retiring much quicker. I'm guessing I'm not the only one this logic applies to.
thanks to laura and declan for their efforts.
I take issue with the tax-and-transfer model, preferring an insurance-and-risk model. Insurance-and-risk makes explicit some important incentive issues and also draws attention to the redistributive effects of a broad range of policies, which do not actually involve the government sending someone a check.
As one example, in the U.S., since WWII, Republican administrations generally achieve a somewhat higher, and more volative unemployment rate than Democratic Administrations. There's no tax and transfer there, but there are effects on the distribution of income. Even a slightly higher unemployment rate helps to suppress wage growth.
Post a Comment
<< Home