White Hills Loan 13minpaydayloans.com White Hills Loan 13minpaydayloans.com

If you are familiar with my writing, you know that for years I’ve been covering the proverbial non-barking dog: the textbook relationship between taxes and economic growth, namely that higher marginal rates make the economy grow more slowly, is not borne out in real world US data. Sure, there are a whole raft of academic studies that claim to show just that, but all of them, without fail rely on rather heroic assumptions, and most of them throw in cherry picked data sets to boot. Leaving out those simple assumptions tends to produce empirical results that fail to abide by the most basic economic theory. This is true for data at the national level and at the state and local level. Making matters more uncomfortable (and thus explaining all the heroic assumptions and cherry picking of data in the academic literature) is that the correlations between tax rates and economic growth are actually positive. That is to say, it isn’t only that we do not observe any relationship between tax rates and economic growth, in general, it turns out that faster economic growth accompanies higher tax rates, not lower ones, and it doesn’t take fancy footwork to show that. A few simple graphs and that’s that.

Now, obviously I sound like a lunatic writing this because it goes so far against the grain, but a) I’ve been happy to make my spreadsheets available to any and all comers, and b) others have gotten the same results on their own. Being right in ways that are easily checkable mitigates my being crazy (or a liar, for that matter), but it doesn’t change the uncomfortable fact that data requires a lot of torture before conforming to theory. And yet, that’s the road most economists seem to take, which explains why economics today is as useless as it is. It also speaks poorly of economists. The better approach is come up with theory that fits the facts rather than the other way around.

I’ve tried a few times to explain the relationship that I’ve pointed out so many times, but I never came up with anything that felt quite right. I think I have it now, and it’s very, very simple. Here goes.

Assumptions:
1. Economic actors react to incentives more or less rationally. (Feel free to assume “rational expectations” if you have some attachment to the current state of affairs in macro, but it won’t change results much.)
2. There is a government that collects taxes on income. (Note – In a nod to the libertarian folks, we don’t even have to assume anything about what the government does with the taxes. Whether the government burns the money it collects in a bonfire, or uses it to fund road building and control epidemics more efficiently than the private sector can won’t change the basic conclusions of the model.)
3. People want to maximize their more or less smoothed lifetime consumption of stuff plus holdings of wealth. More or less smoothed lifetime consumption means that if given the choice between more lifetime consumption occurring, with the proviso that it happens all at once, or a bit less lifetime consumption that occurs a bit more smoothly over time, they will generally prefer the latter. Stuff means physical and intangible items. People also like holding wealth at any given time, even if they don’t plan to ever spend that wealth, because wealth provides safety, security, and prestige, and for some, the possibility of passing on some bequest.

(If the first two look familiar, they were among 8 assumptions I used last week in an attempt to get where I’m going this time around. Note that I added two words to the second assumption. More on last week’s post later.)

Due to assumptions 1 and 3, people will want to minimize their tax burden at any given time subject provided it doesn’t decrease their lifetime consumption of stuff plus holdings of wealth. Put another way – all else being equal, peoples’ incentive to avoid/evade taxes is higher when tax rates are higher, and that incentive decreases when tax rates go down. Additionally, most people’s behavior, frankly, is not affected by “normal” changes to tax rates; raise or lower the tax rates of someone getting a W-2 and they can’t exactly change the amount of work they do as a result. However, there are some people, most of whom have high actual or potential incomes and/or a relatively large amount of wealth, for whom things are different. For these people, some not insignificant amount of their income in any year comes from “investments” or from the sort of activities for which paychecks can be dialed up or down relatively easily. (I assume none of this is controversial.)

Now, consider the plight of a person who makes a not insignificant amount of their income in any year comes from “investments” or from the sort of activities for which paychecks can be dialed up or down relatively easily, and who wants to reduce their tax burden this year in a way that won’t reduce their total more or less smoothed lifetime consumption of stuff and holdings of wealth. How do they do that? Well, a good accountant can come up with a myriad of ways, but in the end, there’s really one method that reigns supreme, and that is reinvesting the proceeds of one’s income-generating activities back into those income-generating activities. (i.e., reinvest in the business.) But ceteris paribus, reinvesting in the business… generates more income in the future, which is to say, it leads to faster economic growth.

To restate, higher tax rates increase in the incentives to reduce one’s taxable income by investing more in future growth.

A couple acknowledgements if I may. First, I would like to thank the commenters on my last post at the Presimetrics and Angry Bear blogs, as well as Steve Roth for their insights as they really helped me frame this in my mind.

Also, I cannot believe it took me this long to realize this. My wife and I are certainly not subject to the highest tax rate, and yet this is a strategy we follow. At the moment, we are able to live comfortably on my income. As a result, proceeds from the business my wife runs get plowed back into the business. This reduces our tax burden, and not incidentally, increases our expected future income.

Added, December 19, 2010. 1:06 PM
While the theory was an attempt to explain the data, I originally thought that I could provide an additional test. That test originally appeared in the post, but I have removed having had a night to think about it. The test provided additional evidence that higher tax rates cause changes in behavior that are contrary to expectations of the standard economic theory (and aligned with my little story) but it occurs to me that I can construct a story line from the standard economic theory that would justify the test’s results as well. Since it did not provide a definitive either/or (i.e., either the standard economic theory is right or my story is right) test (regardless of its results), it had to come out, as did other references to the test. I’ll see if I come up with something else.

While I was added, I made a few more changes for legibility.

24 Responses to “A Simple Explanation for a Strange Paradox: Why the US Economy Grew Faster When Tax Rates Were High, and Grew Slower When Tax Rates Were Low”

  1. Brian says:

    Frame our mind is your signature phrase. That’s not easy. It’s easy to show our economy grows when tax rates are higher. Our economy is better when tax rates are higher. Yes, let’s test the story. Thank you.

    A good country. Our nation is good. I am happy here. I give. I take. I provide. That’s my gift in return. I can be happy when you are happy. I work for that. We all do. We all must. This is America. Our country. A good country. We make America. We are America. There’s no argument about that. You want an end of the day? You want to be right? We make good country. Do our country. Do America.

    See? Motive goes well with analysis too. I like that.

  2. Jeff V says:

    Good morning Mike,

    Interesting column. If I may, I would like to play devils advocate against your idea.

    You state that higher tax rates makes your wife more likely to re-invest proceeds of her business back into her business. This applies in general; afced with higher income taxes, any current business owner would be more likely business owner would be more likely to re-invest business proceeds back into the business, to reduce their tax burden.

    My thought is, what about potential future business owners? Suppose I was considering quitting my job to start a new business, and we will live primairly off of my wife’s income for a while while I get established. Is it possible that high income taxes would discourage me from starting a new business? That is, could a higher income tax encourage investment in exisiting businesses (and thus promote short-term economic growth), but discourage new businesses from forming (and thus damp long-term economic growth)?

    I’m not sure I agree with my argument (I am already forming a counter-argument in my head), but just food for thought.

    - Jeff V

    • Mike Kimel says:

      Jeff V,

      I think most people who form businesses (my wife and sister, and two of my three closest friends are serial entrepreneurs so I see this at close hand) are driven to do it. The income is not the primary thing, its the “doing my own thing” that seems to be the driver. Thus, even if tax rates were higher on business than on a regular job (and to some extent, they are with SS taxes, though businesses have their own tax evading mechanisms as I’ve noted) people who form businesses would still be out there doing it. If tax rates are the same but “too high” (and I don’t want to get into what “too high” is), if its too high to “work for the man” it might still not be too high to work for yourself.

      I dunno. That’s what I got right now.

  3. [...] Grew Faster When Tax Rates Were High, and Grew Slower When Tax Rates Were LowCross posted at the Presimetrics [...]

  4. hacksoncode says:

    It seems that this analysis requires that *eventually* the tax rates come down so that there’s a period in which income can be realized and spent.

    If people think that tax rates are going to continue to be high forever, the time value of money would tend to favor consumption now.

    • Mike Kimel says:

      hacksoncode,

      Actually, it does not. People derive benefit from accumulating wealth for a number of reasons. Think of the elderly billionaires and centimillionaires who will die with billions and hundreds of millions in the bank.

  5. Joe Dillon says:

    Okay Mike, here’s my shot.

    Your model implies instant reaction and correlation between tax rates changing and behavior.

    Does your dataset where you are finding a positive correlation between tax rates and growth have any kind of time offset? Does that offset effect the correlation?

    Example 1 – Rational actors make decisions to build oil rigs based on current economic climates and conditions. These decisions are usually conservative due to the 18 month to 3 year span between order and delivery. Consequently, the decision to build more (or less) oil rigs based on the tax structure could have a significant lag before the results of this decision making become apparent.

    I feel that this would be true for a lot of heavy industry applications (GE, Boeing, Intel) etc. and am curious to see how your modeling took this into account.

    Thanks for the great read!

    • Mike Kimel says:

      Joe Dillon,

      I don’t want to rehash everything I’ve done, but a good summary are some of the more recent posts at the Presimetrics site. I’ve looked at lags of all sorts. I’ve looked at state and local and federal data. A recent summary is this post: http://www.presimetrics.com/blog/?p=253. Note that I am happy to provide my spreadsheets to anyone who wants them, but I also make it easy for you to replicate the results yourself if you don’t trust me.

  6. Joel says:

    Interesting. Studies argue against assumption 2, showing a hyperbolic discounting function which, for a typical person, would kick in fairly quickly as consumption was delayed. I guess it’s possible that a typical dollar value of investment is controlled by a person with an abnormally flat discounting function (the more I think about it, the more likely it seems), but I think there’s an alternate explanation.

    Under the assumptions that 1) a typical employer competes in the market, based on profits minus taxes, with taxes falling to zero if profits are zero or negative, 2) reducing labor costs is widely regarded as a boon to profitability, and 3) there exists some class of persons whose economic activities are limited by their opportunities to earn, we have the following:

    Any set of firms in a given market will include some which spend more on labor, and others that spend less. All else being equal, a firm with higher labor costs will be less profitable than one which spends less on labor. Those firms which temporarily break even or run a loss are in a relatively better position once taxes are accounted for, helping to relieve pressure to cut labor costs, which would limit earning potential and concomitant economic activity.

    This is an especially powerful effect in cases where cutting labor costs would not improve profitability, but employment decisions are made under the assumption that it would.

    • Mike Kimel says:

      Joel,

      Assumption 2 is that the gov’t collects taxes. I think you mean assumption 3. But think of it this way – just about everyone is trying to have assets in their bank account when they die. Billionaires die with billions, they don’t spend or give it away before they die (in most instances). Human beings derive utility from controlling assets.

      As o your story of employers competing – if I understand your point, its that sometimes players aren’t rational and they don’t realize the (after tax) benefits of some of their options. I agree that does happen.

  7. Andrew Jaeger says:

    So is what you are basically implying then is that there is some type of Laffer-esque curve for GDP in relation to the marginal tax rate? Because obviously say if the marginal tax rate were 100%, people would surely be unlikely to work in the first place as they would have no take-home income, and they certainly can’t reinvest all of it into their business as they need some sort of income on their own to live. And if this is what you are implying do you think you would be able to construct the graph from your data, or do you need more/different data to do so?

    • Mike Kimel says:

      Pssst. Don’t spoil the surprise. That’s my next post. And yes, there is a fastest growth tax rate (subject to caveats) and yes, I will point it out on a graph. And no, we aren’t anywhere near close.

  8. [...] This post was mentioned on Twitter by Jens Rasmussen and Dave Spatholt, dennisjordan. dennisjordan said: All tax cuts must eventually be paid for; but do U.S. taxpayers act as though high taxes will always be voted downward? http://bit.ly/ft5pHn [...]

  9. mypetrock says:

    I agree with your conclusion, but not necessarily your assumption in point (3). Your Platonic consumers don’t seem like the type who would be caught in a sub-prime loan with mounting credit card debt. My brother is a personal injury lawyer who works on construction cases. When his clients get their verdict, their first stop is not with a broker generally. It is with a car dealer where they buy the biggest truck they can afford. I would like to see how your model changes when you skew the data to have a significant portion of the bottom tax bracket as distinctly non-Platonic consumers.

    • Mike Kimel says:

      mypetrock,

      FWIW, I’ve been thinking of this mostly as a top tax bracket thing. I could be wrong, of course, but I don’t think this whole story applies all that much at the lower tax brackets. Most people who don’t make some minimum amount of income aren’t going to have investments into which they can roll over income, or the choice as to whether to keep rolling over the income rather than using it for basic consumption.

  10. benkyouburito says:

    Jeff V (directly) and Mike (by extension)–

    Jeff asks “Is it possible that high income taxes would discourage me from starting a new business?”

    I think this question would fade if the nature of taxation is specified as progressive and/or the “higher” tax is specified as a higher marginal tax rate.

    This would be in line with the data as out income tax has always been progressive.

    This reminds me of the argument that is always brought out when one mentions that the marginal tax rate in the 40s – 60s was 91% and our economic expansion was meteoric. The logical gymnastics are impressive.

    “But nobody ever paid that 91% rate” is the usual retort. The implication is that since no one made the $500k necessary to fall in the top bracket that the 91% rate was superfluous. This reversal of cause and effect is necessary if you want to paint high tax rates as harmful to a company.

    So where did all of the profits go during those decades? CEOs didn’t usually earn over $500 because the 91% tax would eat it up and the company could get better value spending that money elsewhere. Corporations don’t pay tax on revenue that is spent on capital investment or payroll. The higher marginal rate forced companies to invest in the most permanent parts of the company.

    This is a great data driven approach to this problem Mike. Thank you.

    • Mike Kimel says:

      benkyouburito,

      You raise some excellent points – the reverse causality that explains ballooning CEO pay is not something I thought of before you mention it.

      The only place I disagree a bit, and this kind of leads me to like Jeff V’s point a bit more that I did before – the mechanism in the post kind of does provide a reason to invest in existing enterprises rather than start a new one.

  11. great_expectations says:

    I am interested as to how this works long term as far as rational expectations goes. In order for this behavior to occur the top income earners must have some rational expectation that future tax rates will decline, allowing them to capitalize on their delayed earnings. If they come to expect a continuously higher rate, then what is their incentive for delaying income?
    Also, this goes a little outside your model, but seeing as top income earners tend to pay a lower net rate as a result of the tax advantage of capital gains earnings, would this increase cause the additional investment to be directed differently?

    • Mike Kimel says:

      great_expectations,

      In the model (and in real life) entities derive utility from gaining wealth, even if they don’t spend it. A lot of very wealthy people die with a lot of assets. It can’t all be attributed to the bequest motive. After all, they could have easily passed on those assets to heirs prior to their death if they chose. Additionally, people die with a lot of assets even if they have nobody to whom they care to pass those assets. Clearly this indicates that people value having wealth they’ll never spend.

      As to the fact that “investors” get charged lower taxes (cap gains) – well, I think that just decreases the incentive to invest in some ways. Say you and a buddy both made $1 billion last year. Each of you always wanted a 757, a custom built Ferrari, and, I don’t know, Guatemala. Now, pull the money out, and you pay a tax rate of 15%, but “reinvest” and you pay zero. Your buddy has a similar deal, but its pull the money out and pay 30% or leave it in and pay zero. Which of you is more likely to let it roll for another year? Now… the one thing that cap gains rate does is that it provides and incentive to those who can “invest” in ways that gives them returns that are subject to cap gains taxes rather than other types of investments.

  12. [...] that enhance economic growth, and in part because raising tax rates, at least at some levels, actually generates more effort from the private sector. However, the benefits of increasing tax rates slow as tax rates rise, and eventually peak and [...]

  13. [...] can be at least partly explained by the relationship between tax rates and investment. As I stated here, in my opinion, higher tax rates can lead to more investment. After all, one way a person who owns [...]

  14. [...] Consider this a companion piece to Why Don’t Tax Havens Become Economic Powerhouses? and A Simple Explanation for a Strange Paradox. [...]

  15. Keating Willcox says:

    1. The reason for economic growth is not new businesses. There is always new and cool stuff to buy. It is demand from customers who have money and decide to spend it.

    2. The reason for economic growth is also a sense that hard work, education, and investment can produce a personal life with more comfort and goodies.

    3. Omitting those government operations which are basically enforced savings such as “free” university education, “free” medical care and “free” retirement insurance, most government spending is considered a penalty against the individual, wealth taken by force to give to some other cause.

    4. Counties like Canada, and states like Texas show that lowering the cost of government produces prosperity, and Europe and the US show that expensive government slows non-financial growth way down. The effect is hidden by the many other factors that can help or hurt growth…

Leave a Reply to Andrew Jaeger


- eight = 0