Relevant and even prescient commentary on news, politics and the economy.

Economic hitman

by Mike Kimel

Cross posted at the Presimetrics blog.

I guess when you’re a very not famous (co-)author like yours truly, people start contacting you with information about their books. I got an e-mail today from another currently very not famous author plugging his book, and I found it to be an interesting concept.

The book is called The Economics of Ego Surplus by Paul McDonnold, and it is a “novel of economic terrorism.” This website allows you to read the first 54 pages of the novel. I personally read about ten or so, and decided to order the rest of the book. (Note – I don’t know Paul McDonnold, never heard of him before he sent me an e-mail, and am getting nothing out of this. He did want to send me, maven that I am, a free copy of the book but I am sending him a check.) FWIW, its not so polished that it doesn’t come across as a first novel, but on the other hand, it smoothly blends in some economics/finance with a Dan Brown-style conspiracy. Put another way – it reads like the bestseller I picked up at the airport a couple of weeks ago before getting on a plane with the added benefit of dealing with a field I find interesting. Put yet another way, it reads like books by Paul Erdman, who I used to read for fun back in college and grad school. For those of us who like our economics/finance, and enjoy the occasional (for me, time is a major constraint these days, and thus we’re mostly talking when I travel) thriller its nice to see another example of the two combined.

Can you think of other examples from this genre?

Tags: , Comments (2) | |

Hauser’s Law is Extremely Misleading

by Mike Kimel

Hauser’s Law is Extremely Misleading
Cross posted at the Presimetrics blog.

A friend sent me a link to this Wall Street Journal opinion piece by W. Kurt Hauser. Who is he, you ask? Here’s what it says at the bottom of the article:

Mr. Hauser is chairman emeritus of the Hoover Institution at Stanford University and chairman of Wentworth, Hauser & Violich, a San Francisco investment management firm. He is the author of “Taxation and Economic Performance” (Hoover Press, 1996).

Before I go on, let me note that in this piece, Hauser masterfully demonstrates the Hoover Institution approach to data. The piece contains enough, er, material that I could write several posts on it. Maybe I will, but for now I want to focus on his key point. Here are the opening paragraphs of the essay modestly entitled “There’s No Escaping Hauser’s Law”:

Even amoebas learn by trial and error, but some economists and politicians do not. The Obama administration’s budget projections claim that raising taxes on the top 2% of taxpayers, those individuals earning more than $200,000 and couples earning $250,000 or more, will increase revenues to the U.S. Treasury. The empirical evidence suggests otherwise. None of the personal income tax or capital gains tax increases enacted in the post-World War II period has raised the projected tax revenues.

Over the past six decades, tax revenues as a percentage of GDP have averaged just under 19% regardless of the top marginal personal income tax rate. The top marginal rate has been as high as 92% (1952-53) and as low as 28% (1988-90). This observation was first reported in an op-ed I wrote for this newspaper in March 1993. A wit later dubbed this “Hauser’s Law.”

Over this period there have been more than 30 major changes in the tax code including personal income tax rates, corporate tax rates, capital gains taxes, dividend taxes, investment tax credits, depreciation schedules, Social Security taxes, and the number of tax brackets among others. Yet during this period, federal government tax collections as a share of GDP have moved within a narrow band of just under 19% of GDP.

OK. So, Hauser’s point is clear – no matter what happens to taxes, the government only manages to collect about 19% of GDP. Presumably then, from a perspective of paying down debt, there’s no benefit to raising taxes and plenty of benefit to cutting taxes. (Later he goes on to argue that lower taxes = faster growth, which I’ve dispensed with in the past – latest example here. Still, if given time, I might come back and examine Hauser’s special way of reaching his conclusion. But that’s for another day.)

Now, they say a picture is worth a thousand words, so let me put up a graph. And for grins, let me embed a small table in that graph. The graph shows total federal receipts divided by GDP. However, it is color coded. In years when there is a cut in the top individual marginal tax rate, or when the most recent change in the top marginal tax rate was a tax cut rather than a tax hike, the area under the curve is colored gray. When there is a tax hike, or the most recent change was a tax hike, the same area is colored red. Here’s what it looks like:

Figure 1.

So there it is. There’s Hauser’s law. Notice the size of his narrow band – its width is over 5% of GDP! Now take a gander at the little table. In tax hike periods, the smallest amount collected was 18.3% of GDP. By contrast, the median collection in tax cut periods is 18.2%; in other words, in over half of the tax cut years, collections were less than the smallest amount ever brought in during the tax hike periods. Furthermore, both the median and average for the two series are a full percent of GDP apart. Hauser is essentially sweeping humongous differences under the table.

Think Hauser doesn’t know this? I don’t. He’s been staring at the data, and using it to make arguments for a very long time. He also writes extremely precisely. At no point does he make a false statement, but I for one reached all sorts of mis-impressions just from his opening paragraphs. Like I said, its a masterful example of the Hoover craft.

Tags: , Comments (42) | |

A Proposed Bet for Professors Bryan Caplan and David R. Henderson

by Mike Kimel

A Proposed Bet for Professors Bryan Caplan and David R. Henderson (and Anyone Else Who Believes Lower Taxes Generate Faster Economic Growth)
Cross posted at the Presimetrics blog.

Professors Caplan and Henderson,

Both of you have had recent posts that indicate you have some enthusiasm for betting on economic outcomes. (Your co-blogger at Econlog, Arnold Kling seems less enthusiastic about bets, and thus I have not addressed him by name here.) I have a few criticisms of your approach to betting. The first is that, frankly, y’all are betting on some rather peripheral issues. Why not cut to the chase? Why not propose a bet on something vital to your way of thinking but with which many people disagree? For example, as libertarians you believe that lower marginal tax rates on the “producer class” result in faster economic growth in a well-functioning, more or less market based economy, and that this outcome can be observed in the US economy. (Forgive the wordiness, but I want to be precise so you don’t think I’m trying to trap you up in a technicality or some oddball example.) I believe you are generally wrong, at least about the US economy. Many people share your beliefs, and many people share mine, so this would be an ideal topic on which to bet if your goal is to prove a point.

Another criticism I have of the bets I’ve seen you propose is that your bets tend to be based on a small number of events, typically one or a handful of observations occurring over ten years or less. But that is too short a period to leave out the effect of random fluctuations, acts of God, or long running conditions. For example, though I haven’t verified the data myself, I understand that it has been pointed out that had the Julian Simon bet (a favorite of Professor Henderson’s) occurred a few years later results would have been different. A truly fair bet would look at more data. In fact, an ideal bet would look at many different overlapping long time periods. Results over ten year windows, twenty windows, thirty year windows, etc., would all combined to ensure that the results aren’t just an artifact of the data.

Another safeguard which helps get at a “true outcome” rather than some random fluctuation is to consider whether the effect you are looking at can have lags of different lengths. For example, it may be that the marginal tax rate in 2010 might affect growth rates from 2009 to 2010, or from 2010 to 2011, or from 2010 to some later year. After all, as any libertarian would say, if you pay less in taxes this year, it means more money in your pocket this year. Since you spend more efficiently than the government, that creates more growth this year, and that additional growth has positive effects next year too. Of course, at some point, the future effects of today’s tax rates dissipate. Not having a precise theory, it probably pays to consider several of these “effect periods” to (perhaps) coin a term.

The third problem I have with your bets is that, frankly, it takes too long to find out who won. Professor Henderson indicates in one post that he’ll probably be settling up with the estate of fellow bettor. If bets are intended as a way to help move the field, not to say the bettors beliefs, forward, results have to come in more quickly to make an appreciable difference. Now, at first glance, this last complaint kind of clashes with my previous criticism that ten years of data is just not enough. But if you think about it, there’s an easy way to square the circle: the obvious solution is to bet on outcomes that occurred on the past.

Now, before you say that’s silly, hear me out. You wouldn’t (dare I say couldn’t?) be a libertarian if you didn’t believe that historically, economic growth in the US was faster when the marginal income tax rates on what you would term the productive class were lower than when they were higher. And in the unlikely event you’ve read anything I’ve written, whether on the Presimetrics or Angry Bear blogs, or the book I cowrote with Michael Kanell, you would be aware that I am pretty certain the US economy is not characterized by such a relationship between marginal tax rates and economic growth. Simply put, what each of us knows about the past contradicts what the other knows. At least one of us has to be wrong.

Given how bet-happy you all are, and your core beliefs, I would have expected you to propose this one (not necessarily to me, who you no doubt don’t know from Adam) a long time ago. To be precise, here’s the bet I would have expected you to issue:

For the vast majority (say, at least 70%) of overlapping windows, the correlation between top marginal individual income tax rate and the growth in real GDP will be negative. Windows of data to be considered are ten years long, twenty years long, etc., through sixty or seventy years long. Growth rates to be compared with marginal tax rates at time period t include t – 1 to t, t to t+1, t to t+2, t to t+3, t to t+4, and t to t+5.

Now, I could see variations of that bet. For instance, Professor Henderson has indicated in a recent working paper that he doesn’t believe National Accounts data for the WW2 years are accurate, so perhaps he would structure the bet to only use data from 1946 on, rather than the 1929 on which is possible with the official BEA data. Alternatively, perhaps t to t+5 might seem a little much to consider, or perhaps one would prefer to include t to t+x where x is something larger than 5. Nevertheless, this is very close to the bet I would have expected to be proposed from people with strong libertarian beliefs who like to engage in wagers on economic outcomes.

One other thing – notice that I indicate the correlations should be negative well over half the time. I have yet to hear a libertarian hedge when he/she tells me about the benefits of lower taxes. Getting a touch above 50% just doesn’t fit that with that sort of certainty, and is more akin to random fluctuation, is it not? (But don’t worry, where we’re going, the distinction between 50.00001% and something more appropriate to your level of certainty won’t matter.)

What is left is to consider – what should have been the size of the bet we should have expected to see. Now, Professor Caplan has recently noted:

But why are small sums enough to deter 95%+ of the people who disagree with me? I see two main reasons:

1. We aren’t just betting $100. We’re betting $100 plus reputation plus bragging rights. That’s why I prefer to bet the famous. The Simon-Ehrlich victory wouldn’t have been nearly as awesome if Simon bet a random Malthusian.

2. Many spouses, perhaps most, disapprove of betting. They think you’re irresponsible when you bet, and stupid when you lose. Imagine how badly they’d react if the stakes were $25k! Even the victor might find himself stuck in the doghouse.

So… $100 plus bragging rights is about right. Of course, I’m not famous, so I doubt the bet would have been issued to me. I would have taken the $100 bet, though, if offered. More – well, probably not, despite my certainty, given item number 2. Nevertheless, I am surprised that neither of you offered this bet to someone.

But here’s the thing. You would have lost that bet. And we’re not talking by a smidge, we’re talking by a country mile… or seventeen and a half.

Here’s what I get:

by Mike KimelFigure 1.

(Note – you might have to click on the figure to see it in full. It seems to cut off on my browser. The same is true of the next figure.)

The way to read this graph…. consider the cell with t to t+3 on the horizontal and 50 years on the vertical. That cell has 62.1% in it. That indicates that of the 29 fifty year windows in which you can measure the growth in real GDP from a given year to three years later, 18 of them (or 62.1% of them) show a positive correlation between the top marginal tax rate. That is to say, in 62.1% of those windows, growth is faster when top marginal tax rates are higher than when tax rates are lower.

Notice… most of the squares have numbers above 50% in them. That means, in most situations we considered, more often than not, the correlations between marginal tax rates and growth rates are positive, not negative. When the negative correlations do occur, they tend to occur over the very short term. Put another way – they have negative repercussions that hit later. (And yes, that is what the table indicates.) Over longer periods of time, the percentage of time positive correlations are observed approaches 100%. This cannot in any way be reconciled with libertarian theory.

FWIW, the table above represents a grand total of 1,652 observed correlations between the top marginal tax rate and growth rates of real GDP. 56.5% of those correlations are positive.

Note… I haven’t included it in the table, but for giggles I checked the t to t+10 results. For ten and twenty year windows, the percentages are below 50%. For thirty year windows and up, the percentages are above 50% and go above 70% at 40 year windows and hit 100% at the 70 year windows. Put another way… t to t+10 looks an awful lot like t to t+4.

Now, say you’re Professor Henderson and you want to discard the data through the end of WW2. In that case, you come out looking even worse:

Figure 2.

Now, 64.6% of all the correlations observed are positive.

Now, this post is starting to get awfully long, so let me wrap it up. I think you should offer this bet. In fact, my advice to any libertarian or conservative is to offer to make this bet. Sure, its easy for me to say, because the bet goes against you, but I promise if you offer the bet or something similar I will refrain from jumping in so I’m not making that suggestion for personal gain. The reason I think you should offer this bet is that, knowing you’d lose gives only a few options:

1. You can change your beliefs.
2, You can tapdance into the opposite result. To some extent, that’s where the economic profession is now. There are any number of studies by well known academics that show that cutting taxes lead to faster economic growth under some or most conditions, and they all require either weird special cases or assumptions that, frankly, could be used to show that a 400 year old sketch of a chicken is a nuclear submarine.
3. You can pretend none of this ever happened.
4. You can show there is a problem with what I have done or proposed.

Now, its possible I’ve made a mistake, but to repeat myself, if you’ve read anything I’ve written before, you’ll find that I’ve been on a “the data shows that lower taxes do not equal faster economic growth” kick for a long time. I’ve gotten here every which way, using data from all sorts of sources and at all levels of granularity. In this case, I’m guessing that if you included windows of 11, 12, 13, etc. years, you might push the percentage of positive correlations down. For all I know, with judicious fiddling, you might even get to a point where a slight majority of cases have a negative correlation. I don’t have an institute or a university paying me to make this sort of argument and I’m running out of spare time this afternoon. But even if you got that percentage down a bit – the libertarian position is not that lower taxes lead to faster economic growth somewhere around half the time, is it? And frankly, it would take a heck of a lot to get that number down for a Henderson post-WW2 look. And no matter what, you aren’t going to escape one more detail – over longer periods of time, the correlations are overwhelmingly positive. I’d hate be touting the benefits of lower taxes and having to explain that fact.

Moving on, the problem with option 2, the status quo, or option 3, is that its simple enough to show what I’ve shown. The results are there. As noted above, I’ve done this sort of thing so many times, so many ways, with so many different data sets, and at so many levels of granularity. Sooner or later someone that other folks do listen to will discover the same thing. Then what?

As to option 1, well, Upton Sinclair said it best a long time ago, “It is difficult to get a man to understand something when his job depends on not understanding it.” And frankly, its hard to see GMU or the Mercatus Institute or Hoover or even the blog where you write keeping you on if you start telling people that higher top marginal rates are correlated with faster economic growth. You have a lot to lose if you change your beliefs.

So if you can’t take any of these options, you really need a different approach. And what’s better than going on the offensive? Offer up the bet. Sound confident- a true believer would insist that correlations between lower taxes and faster growth should be there 90% of the time, right? Heck, issue odds. Do that and people might assume you know the results favor your position. People are lazy, and they don’t check. That’s why so many people believe so many things that simply don’t hold up when confronted by data.


Mike Kimel

PS. The Excel file containing the data and analysis that went into this post is published as a webpage here. I’m not quite sure why but the ten year results seem to have acquired an error upon uploading into google. Everything else seems OK, but should anyone want the original Excel file, drop me a line at mike period and my last name, all at gmail dot com.

Tags: , , Comments (25) | |

E Pluribus Unum and Our Finest Hour

by Mike Kimel

E Pluribus Unum and Our Finest Hour
Cross posted at the Presimetrics blog.

But if we fail, then the whole world, including the United States, including all that we have known and cared for, will sink into the abyss of a new dark age made more sinister, and perhaps more protracted, by the lights of perverted science. Let us therefore brace ourselves to our duties, and so bear ourselves, that if the British Empire and its Commonwealth last for a thousand years, men will still say, This was their finest hour.

Winston Churchill, Speech to the House of Commons, June 18, 1940

Assuming, as many do, that the British Empire ended some time around the handover of Hong Kong, it did not last a thousand years. (Britain and its Commonwealth, of course, are still going strong.) Nevertheless, I suspect many would say that Churchill got his way, and that the Battle of Britain and the remainder of World War 2 was, in fact, the finest hour of the British Empire.

What about the American Empire? If we define that institution as existing from some time around the Spanish American War (1898) to the point where it was overextended and became unable to impose its will on friend or foe alike (i.e., some time around 2005), what was its finest hour? What were its most impressive achievements, those that will be written up in history books a thousand years from today?

I am not a historian, but I have a few guesses, in no particular order: (below the fold)

1. Serving as the arsenal of democracy in WW2
2. Putting people on the moon and bringing them safely back to earth.
3. The development of mass media and long distance communications.
4. Almost eradication of polio (yes, a worldwide effort, but just about every significant piece of the project was done in the US)
5. The Green Revolution (a little less US-dependent than the polio effort, but US entities played the biggest role)
6. The Manhattan Project and the development of nuclear energy
7. The early development of genetic engineering (I suspect US dominance in this field will be ending very soon)
8. Construction of the Panama Canal
9. Airplanes
10. An automobile in every driveway
11. The electric grid

I’m sure I’m forgetting something important, and there are, no doubt, things we regard as small that will be viewed as important one day. Still, I would be surprised if what is eventually viewed as the greatest achievement of the American Empire is not on that list. However, not all of the items listed will survive the test of time. Some will be forgotten, some may prove more or less irrelevant over the long haul, some will come to be viewed as a facade and some will be decried by our descendants. Still, its probably not a bad list, and I think its good enough for the purpose of this post, which is to note: the role of the private sector tends to play a relatively small role when it comes to the big achievements. Furthermore, the piece of the private sector that contributes the most to the big advancements, the ones that will be remembered, is the not-for-profit piece of it.

With the possible exception of 3, 9 & 10, the for-profit private sector played the role of sidekick or supporting actor. The main role, the driving force, the entity that either provided the original vision and/or drove that vision through to completion was the government, with much of the remainder provided by academia (heavily funded by the state whether public or private) or NGOs. But even where the private sector led the charge, the government’s role was huge. Henry Ford may have revolutionized the production of cars, but without the government producing roads (not to mention the freeway system), their development would be limited to where they could be used for local transportation. Most of the big achievements, and, I am comfortable making this statement, the finest hour of America, whatever that is judged to be a thousand years from now, are driven by government policy, government actions and government grants.

Why is that? After all, the private sector, after all, makes up the biggest chunk of the economy. Size alone doesn’t isn’t enough to create achievement – the most significant achievements in the private sector usually aren’t those produced by the biggest companies. Similarly, its hard to construct a story that involves the government coming into a field and bigfoots over the early efforts of the private sector. Instead, the government is providing a role that the private sector simply isn’t, cannot, and will not. Why? I have a guess. I suspect it comes to the profit motive. Projects of this nature are risky and costly and hard to make money off of for a very long time, all of which are factors that discourage the private sectors. But the private sector has another problem with “finest hour” type accomplishments, which is evident when you think of Britain and Churchill’s speech. Britain may have been, to Hitler, little more than a “nation of shop-keepers”, but those shop-keepers were willing to fight for an idea, a cause they all had in common. However, its hard to imagine a company providing a vision that unites a nation. Occasionally, a company is able to inspire its employees to greatness – think Hewlett Packard before Carly Fiorina and the era of continuous layoffs. However, even then, the reach of its vision, its ability to bring others on-board, is generally limited to that company itself. This is due to human nature. The geniuses – the Einsteins and Borlaugs and Salks aren’t in it for the money, and the rest of us aren’t going to get the warm and fuzzies from increasing the profits of a company for whom we don’t work and in which we don’t own stock.

The only force that can unite the country, that can create a cause around which everyone will rally around, and then only certain circumstances, is the government. E pluribus unum. But that is why the American Empire has been petering out. We are less than two months shy of thirty years from the day when Reagan told us the government is the problem, and we have bought into that mantra hook line and sinker. And in the Tea Party era, it is hard to see how that can be turned around. The long, slow decline is becoming inevitable.

Tags: Comments (65) | |

Extrapolated September 2010 Debt, by President

by Mike Kimel

Extrapolated September 2010 Debt, by President

Cross-posted at the Presimetrics blog.

This one is quick and dirty because I’m very low on time…. Anyhow, these days there’s a lot of talk about the debt, and some talk about how irresponsible Obama is, or maybe its GW. Who knows, there’s a lotta talk and very little hard facts, much less context. So, from working on Presimetrics, I had some debt data lying around and started with that.

The table below shows the total national debt and ending national debt for each President for December of the year before he took office and the growth rate over that time period. For grins I threw in the President’s extrapolated September 2010 debt. That was computed by taking the debt in December (in September 2010 dollars) of each President’s last year in office, and, assuming the rate at which debt had increased during his term would continue all the way to September 2010.

All data is in September 2010 dollars.

To interpret:

In December 1980, a month before Reagan took office, the debt was 2.354 trillion (Sept 2010 dollars). A month before he left office, in December 1988, that debt had increased to 4.866 trillion, which is an annualized growth rate of 9.50% a year. Starting with 4.866 trillion in December of 1988, and increasing at a rate of 9.50% a year would give you 35 trillion and change by September of 2010.

A few things to note… the two Presidents who added to the debt at the quickest rate were GW in first place and Reagan in second. They were followed by Ford, and then Obama, with GHW Bush not far behind. Now, I’ve been pretty critical of Obama for continuing GW’s policies (see Presimetrics, the book I wrote with Michael Kanell, and this) but all in all, as lousy as he’s been, he’s far, far from the worst perpetrator when it comes to fiscal irresponsibility. (And please, spare me the whole “the banks needed saving” when so did many businesses and households… which weren’t saved. I’d be less inclined to carp if the money was spent on keeping Main Street afloat rather than seeing so much flow to Wall Street.) I wonder how the Tea Partiers would react to that information, and whether they are are angrier at GW and Reagan than they are at Obama. Somehow I doubt it.

Note – the data, data sources, and analysis used in this post are available here.

Tags: , , Comments (43) | |

Post WW2 Private Investment v. New Deal Private Investment

by Mike Kimel

Post WW2 Private Investment v. New Deal Private Investment
Cross posted at the Presimetrics blog.

I had a post the other day (which appeared at the Presimetrics blog and Angry Bear, and which was followed up by my fellow Angry Bear, Spencer, here) looking at a paper by David R. Henderson about the supposed post-World War 2 economic boom.

I noted that his view fit into a libertarian/conservative story line, but required not only assuming the GDP (or GNP) data from WW2 is wrong, but also that the data at least through the early ’50s is wrong too, despite the fact that the data fits other known facts pretty well. By contrast, Henderson’s story conflicts with known facts in a number of places.

However, there is one point – another libertarian/conservative myth which comes up in the paper that I’d like to focus on in this post. Henderson tells us:

Why did the economy do much better after the war than at the beginning? We can’t know for sure, but the most likely explanation is the change in administration from Roosevelt, who championed central government planning of the economy, to Truman, who was much less inclined to support government control.

Also see Econospeak’s Prof. Rosser on 1920-21 recession


Before the United States entered into World War II, the New Dealers—the faction of Franklin Roosevelt’s administration that was most hostile to economic freedom—had significant power. During the war, they were largely displaced by more pragmatic people who were not hostile to free markets (thus the quote from Henry Stimson at the beginning of this section).

Moving on…

Roosevelt’s death cleared the way for President Harry Truman. Although he was a New Dealer, Truman had no love for “the long-haired boys” who were associated with the most anti-market parts of the New Deal—people such as Ben Cohen, William O. Douglas, trust-buster Thurman Arnold, price controller Leon Henderson, and Felix Frankfurter. In 1945 and 1946, Truman got rid of a number of New Dealers, including two of the most prominent ones: former vice president Henry Wallace and Harold Ickes.28

Higgs points out that the polling data bear out the perception of a regime change under Truman. As a result of the change, writes Higgs, “Investors were then much more willing to hazard their private property than they had been before the war, as both survey data and financial market data confirm.”29

And invest they did. As table 2 shows, gross private domestic investment in real 1964 dollars was $44.4 billion in 1941. For all the war years it was half or less of that 1941 level. In 1946, it shot up to $51.7 billion, grew slightly to $51.8 billion in 1947, and then grew to $60.6 billion in 1948.

So essentially, Henderson’s belief is that there was a boom after WW2 and that it was caused because greater economic freedom encouraged more private investment. We’ve already dealt with the supposed boom, but what about private investment? Was private investment really booming in the post-WW2 era relative to the pre-WW2 era? Simply put, no, as is evident from the following graph, constructed using data from NIPA table 1.1.6:

Figure 1

From the graph, its pretty obvious that the New Deal easily beat Henderson’s post-WW2 boom when it comes to encouraging private investment. The explanation for why is obvious to anyone who has not bought into libertarian or conservative beliefs about how the economy works.

Tags: , , Comments (51) | |

Discouraging Greg Mankiw From Working Would be Good for the Economy

by Mike Kimel

Discouraging Greg Mankiw From Working Would be Good for the Economy
Cross posted at the the Presimetrics blog.

A few days ago Greg Mankiw had an op ed piece in the NY Times talking about how even small increases in the marginal tax rate would keep him (and by extension, other talented folks like him) from working.

For a laugh, I pulled data on the top marginal tax rate from the IRS and real GDP per capita from the BEA’s NIPA tables. Data on the latter goes back to 1929, and the tax rate info goes back further.

It turns out that the correlation between the tax rate in any given year and the growth rate in real GDP per capita from that year to the next is small but positive. That is, higher top marginal tax rates don’t seem to reduce real economic growth. Look at the tax rate and the annualized growth rate in real GDP per capita for two years, or three, or four, or five, or six (which is as far as I went) and ditto – the higher the marginal tax rate, the faster the economic growth over the next X years. The correlation is positive, if small.

Now, you may be saying to yourself – sure, but the world is very different now than in 1929 or 1942 or 1968. What about in recent times? So let’s start in 1981, which is more or less when a) the ideas that Mankiw endorses took hold and b) Mankiw’s career began. Here’s what that looks like:

Figure 1.

Hmmm… a clear positive correlation between the top marginal tax rate and the growth in real GDP per capita over the next four years. Using three or five year lags decreases the correlation slightly but results are about the same. It would seem that discouraging Mr. Mankiw from working would be a very, very good thing for the economy. That shouldn’t come as a surprise to you if you’ve read my book and maybe I’ll write a bit more about this when I get a chance.

However, while it may seem like I’m being facetious about how discouraging Mankiw from working would be good for the economy, I really believe it. After all, Mankiw’s work consists in large part of advocating a position in his books, lectures, op eds, and as an advisor that, as the graph above shows, is consistent with slower economic growth, and he’s very good at what he does.

Tags: Comments (9) | |

Corporate Tax Rates and Unemployment, A Correction

by Mike Kimel

Corporate Tax Rates and Unemployment, A Correction

Cross posted at the Presimetrics blog.

I’m embarrassed. I messed up last week’s post which appeared at Angry Bear and on the Presimetrics blog. Essentially, I copied in some of the data on tax rates incorrect for three years in the late 1970s; the IRS source I was using was not in spreadsheet form. The error was spotted by JzB. To be honest, I kinda shouldn’t have written that post… it was evident even from the way I started it:

I’ve been kind of swamped, low on sleep, and doing a few book related things in my few waking hours that don’t work or parenting (buy a copy of Presimetrics!!!!), so posting has been light. But I thought I’d do a quick and dirty post today about a hot topic

OK. I’ve gotten a bit more sleep (not enough, mind you, but some), and while the baby is in my care right now, he’s quiet for the time being, so let’s get this done. First off, the topic… I want to look a the relationship between corporate tax rates and unemployment. Links to the data are posted below, but all the information I use, plus the analysis itself, sits in this google spreadsheet.

So, let’s begin.

Figure 1 below shows the data used in this post; the top corporate marginal tax rate (obtained from the IRS) is on one axis and the unemployment rate for individuals sixteen years and over (from the Bureau of Labor Statistics) is on the other axis.

Figure 1

Where we go from here was best explained in the last post:

Now, consider the correlation between the top corporate marginal tax rate and the unemployment rate. If it is true that lower taxes = lower unemployment, the correlation between the two series should be positive. A positive correlation means the series should move more or less in the same direction; as tax rates rise, unemployment rises, and as tax rates fall, unemployment falls.

If the correlation is, in fact, negative, that means that lower unemployment tends to happen when tax rates are higher. Correlation may not be causation, but it would be very hard to argue that cutting taxes on corporations leads to lower unemployment if we do not see a positive correlation between the two series.

Now, obviously, it may take time for tax rates to do whatever magic they might have. So Figure 2 looks at the correlation between the top corporate marginal tax rate and the unemployment rate in the same year, the unemployment rate the next year, etc., all the way through ten years out. Its really hard to see how the effect of tax rates should last beyond a couple of years, but I figured I’d be thorough and put up the figures. I’ll take a pass at interpreting them, but feel free to reach your own conclusions.

Additionally, because whatever effect tax rates might have on unemployment might change over time, each correlation is computed several times: once for the entire 1948 – 2009 sample, a second time for 1960 – 2009, a third time for the period since 1970, a fourth for the period from 1980 and a fifth time for the 1990 – 2009 period. (I didn’t look at just post-2000 because the top corporate rate has been frozen during that period.)

Figure 2

So what do we make of this? As I stated last week, I don’t think the outyears are all that relevant; I don’t think unemployment rates are affected by corporate tax rates ten years earlier, but I included them so no-one accuses me of cherry picking. If you think they matter, explain how in comments. Here’s what I’m seeing in the graph:

1. For the longer samples (1948 – 2009 and 1960 – 2009), the correlation between tax rates and unemployment rates is very close to zero.
2. For the three shorter time samples, the correlations are very big (positive for 1970 – 2009 and 1980 – 2009, negative for 1990 – 2009).

That seems to indicate that corporate tax rates did not use to have an effect on unemployment, but in recent years, they may be never used to have an influence on jobs. That influence may not be in the direction that tax cut proponents keep telling us about, though, as evidenced by the 1990 – 2009 series; anyone who has read Presimetrics won’t be at all surprised, as we point out the same thing in a completely different way. I think its because over time, our economy has become more loop-hole oriented than doing things oriented.

Your thoughts?

Tags: , Comments (15) | |

Hey, Big Spenders… (Status Report on Presimetrics)

by Mike Kimel

Update: Rdan here Take the quiz now…hey big spenders

Status Report on Presimetrics

The Intelligence Report column in Parade Magazine, the newspaper that appears in Sunday newspapers across the country, wrote a short quiz based on Presimetrics. It will appear in this Sunday’s issue.

Michael Kanell and I will log in to the quiz’s site a few times on Sunday and (if all goes well) once more Monday to answer questions anyone might have about the book.

Meanwhile, I just got a copy of the book, and it looks very nice. I believe they’ll be shipping copies of the book to bookstores within days. Additionally, the publisher is working on a number of other media appearances.

submit to reddit

Tags: Comments (5) | |

Tax Burdens and Economic Growth – Answering the Objections

by Mike Kimel

Tax Burdens and Economic Growth – Answering the Objections

This piece is cross-posted with the Presimetrics Blog.

Consistent with findings in Presimetrics, the book I wrote with Michael Kanell which will be released in August, I’ve had some posts whose results contradict standard economic theory. In some cases, readers have insisted that the results must be some sort of anomaly. Perhaps the biggest offender is a graph which appeared in this post. The graph shows growth rates in real GDP per capita by Presidency, where each President is color coded by whether he increased the tax burden (in this case defined as federal government revenues as a share of GDP) or decreased it:

Figure 1

The graph shows that Presidents who cut the tax burden produced slower growth, on average, than Presidents who increased the tax burden. For many people, this doesn’t make sense at all. (An explanation for why these results show up is provided here. Now, I’ve already some of the objections that have been raised to this, mostly in private or in comments at the Angry Bear blog. In fact, the graph above by itself answers one of the objections – I was told many times that FDR only produced faster growth because the economy accelerated during World War 2. Thus, the graph only shows FDR’s results through 1938, which avoids the war, the build-up to the war, and even Lend Lease.

Another criticism is that somehow the fact that I’m assuming that Presidents have an effect on the economy is the problem, and that a better way to do this is to look at the business cycle. But I’ve had posts looking at the business cycle, and increasing tax burdens doesn’t look all that much better as an economic strategy across the business cycle either. For instance, consider the following graph, one of several fairly damning graphs I posted here and here:

Figure 2

If cutting the tax burdens is the right prescription during a recession, it isn’t evident from the above graph.

I’ve also been accused of cherry-picking, though I’ve gone as far back as the data allows, and on most posts on this subject, I’ve actually chucked the WW2 years. In addition, I’ve noted that throwing out outliers doesn’t change results materially. But what if the last eight decades have all been an anomaly? After all, shouldn’t we consider the “Roaring ‘20s,” a period which every textbook tells you was a period of rapid growth. After all, such intellectuals as Glenn Beck, Thomas Woods, and Amity Shlaes are quick to assure that the prosperity of the 1920s was due to tax cuts.

Unfortunately, the NIPA tables don’t go that far back, so we don’t have actual data on real GDP per capita (or the tax burden). But I’ve done my best to examine that claim; the following graph, which first appeared here, shows the top marginal tax rates and the periods (shown in gray) when the economy was in recession.

Figure 3

While the 1920s was, indeed, a period when marginal tax rates were reduced, it was a period of recession followed by recession followed by recession followed by the Great Depression. The longest consecutive months spent outside of recession during the 1920s was 27 months! Two years and a quarter. In fact, the economy was in recession (if not outright Depression) a full 44% of the time during the 1920s. If these were roaring years, those roars were extremely short-lived.

Which moves us to the next issue – I’ve been criticized for focusing on the tax burden rather than the marginal tax rate, the exception being in periods where the tax burden is not available. Frankly, of all the criticisms, that one seems especially weak. For those with enough income, the marginal tax rate is as much a fiction as the MSRP on a car; Warren Buffett’s offer of $1 million dollars to anyone on the Forbes 400 list who could prove they pay a higher share of their income in taxes than their secretary remains untaken.

Moving on to the next criticism… I’ve been told my interpretation of the above graph is wrong – it is not so much that higher tax burdens are correlated with faster economic growth, but rather that administrations that produced rapid economic growth tended to feel they could raise tax burdens, and that administrations suffering from poor growth kept tax burdens low to try to remedy a bad situation.

I dealt with that issue in recent posts by grouping the periods from 1929 to the present into eight year administrations, where possible. Those administrations were made up of the eight year terms of Presidents who served two terms (or in FDR’s case, the first two terms only). Added to those were eight year periods in which a Vice Presidents took over for a President who died or otherwise left office. For each of those eight year terms, I created the following graph . It shows the change in the tax burden in the first two years of each administration along one axis, and economic growth on the other axis:

Figure 4

Clearly, economic growth in years 3 through 8 cannot explain changes in the tax burden in earlier years unless one assumes time travel, clairvoyance, or one heck of a coincidence.

And speaking of coincidence, we arrive at another common complaint: there simply are not enough observations to reach a conclusion of any sort. Left unstated, of course, is that though the tax cutting Presidents had the lousiest economies, and that (as per graph 4) tax cutting led the economic growth, somehow a supposed lack observations validates the idea that lower taxes does produce faster economic growth.

Now, those who complain about the lack of observations generally insist that a) I don’t know the first thing about statistics and b) you need 30 or more observations to reach a conclusion. Now that critique dates back at least three years, having been made by an anonymous blogger for the Economist who I understand is now known as Megan McArdle, and my answer (to point b.) back then is as good of an answer as any:

Which is what degrees of freedom are for… Maybe there’s something wrong with the textbooks on my shelf, but the t distribution tables in the back of those textbooks have as few as 1 (one) degree of freedom. When the degrees of freedom are low, the t-statistics has to be really high in order in to reject H0. Or something like that – what do I know?

Allow me to explain. If there are a large enough number of observations to work with, it is possible to find a statistically significant difference between two things (events, peoples, outcomes, whatever) that at first glance or from a distance look very similar to the unaided eye. However, if there are a very small number of observations, then differences have to be larger and more obvious for them to be statistically significant. Consider three medications to extend the lives of patients with a specific type of cancer, where two have obtained FDA certification and the third was cooked up by the creepy guy who lives two houses down and uses cat puke as an active ingredient. It might take hundreds of observations to tell which of the first two medications is more effective, but it shouldn’t take very many to tell how well the third one compares.

And while in the past I sometimes decided to answer this question by running a t-test or some non-parametric test, it always seems to lead to questions about the assumptions by people who clearly never ran a hypothesis test in their lives but have one or another political point to defend to the death. So let me try something else – an argument by analogy. Consider the graph below.

Figure 5

If such a graph came from a study comparing outcomes of medications, where patients were assigned a medication but otherwise told to go about their daily business, there’d be no argument that, at least as a first approximation, there’s no reason whatsoever to assume that Medication 2 was more efficacious than Medication 1. And if the disease in question was a particularly rare one, and the graph above represented the testing performed on every known sufferer since 1929, most of us would be appalled if a doctor decided to treat the next sufferer with Medication 2. At the very least, we would assume that the burden of proof going forward should lie not with the proponents of Medication 1 but rather with the supporters of Medication 2, whether we understood the mechanisms by which either of these pharmaceuticals worked (or purported to work) or not.

And yet, this graph is identical to Figure 1, except that I changed the title and some labeling.

But there is another problem with the small sample objection. See, there are many ways to test whether there has been a negative correlation between tax burdens and economic growth, and looking at the national economy is only one of those ways. I’ve also had a number of posts at the Angry Bear blog looking at how states have fared over the years, and there are 50 of those. In fact, that was the topic of the first post I ever wrote four years ago next month. In that post, a comparison of the states, using data from 1990 to 2005 yielded the following conclusion:

Thus, the data doesn’t seem to support the idea that lower taxes are associated with faster growth rates. In fact, the opposite is true, especially for the fastest growing states. One way to interpret this is to conclude that taxes are actually below their optimal rates, and therefore, at the margin, the government is actually more efficient than individuals at converting its spending into growth. Society needs a certain amount of public goods (infrastructure, public health, confronting the Canadian menace, etc.) for businesses to thrive, and perhaps we currently have too little provision of public goods rather than too much.

And the other posts I’ve had using similar state level data have all led to the same findings.

Another problem frequently brought up is that growth is too complicated to be explained by a single variable. We agree, and in the book, we actually provide a model that uses several variables. But be that as it may, it isn’t reasonable to state that while cutting tax burdens produces faster economic growth, that effect gets swamped by opposing forces when you look at the data systematically, whether you’re looking at the performance of Presidents, at the growth during business cycles, or the performance of states. Clearly, if reducing the amount that people pay in taxes is so beneficial to the economy, somewhere that effect would show up. It shouldn’t be overwhelmed by other variables pushing in the opposite direction every time one tried to test it systematically and consistently.

Moving on, we have another little gem – that the performance of the two regimes (tax cutters and tax hikers) is not independent. The argument is this: tax hikers do well because they follow tax cutters who laid the foundation for growth. Tax cutters do poorly because they have the misfortune of following tax hikers who set up the economy for a fall.

This one is particularly easy to hit out of the park. First, note that there is only one tax hiker that is followed by another tax hiker: LBJ followed JFK. And LBJ produced the second fastest growth in our sample, which is to say that simply following a tax hiker is no guarantee of poor performance.

Now, look what happens when you consider only Presidents that followed their tax burden cutting peers:

Figure 6

Notice that following a tax cutting President doesn’t mean one will turn in a poor performance… unless one is also a tax cutting President. In fact, tax cutting Presidents that followed other tax cutting Presidents did worse than tax cutting Presidents who followed tax hikers. Imagine that. It’s almost as if the longer tax burdens are cut, the worse the outcome.

And yes, I included Hoover in the above graph though we don’t have the data to know with certainty that his predecessor, Calvin Coolidge cut the burden since Coolidge is renown as a small government guy. But leave out Hoover, and leave out Obama’s first year, and you still aren’t left with anything other than: In fact, tax cutting Presidents that followed other tax cutting Presidents did worse than tax cutting Presidents who followed tax hikers.

Which leads to the sorriest objection I’ve heard, namely that the American public, the constantly gulled American public, has the ability to reason out the outcome of economic policies on the macroeconomy to near-perfection, at least in 4 year increments. And the way it manifests itself here is this: when the economy is about to sour, we elect tax cutters, who, in turn, manage to limit the scale of the impending disaster.

This, ahem, theory (gurgle, choke) is the efficient market hypothesis on LSD. But it has the advantage of being able to explain pretty much anything. The problem is that it does so by breaking everything down to utter nonsense. For instance, it would indicate that the recent housing bubble and economic meltdown, rather than being a surprise, was actually anticipated on some unconscious level by the American public, and selected for as being much better than the alternative. Ditto the Great Depression. So what was this worse thing that was avoided? Locusts? Famine and pestilence? Billions of furious yetis descending on us from their Himalayan stronghold?

And yet, despite the fact that this story makes a virtue out of nonsense, it still isn’t internally consistent. For instance, if the American public understood that only a series of tax cuts were going to save us from something worse than the Great Recession, then wouldn’t GW have managed to have won the popular vote in 2000 and achieved a landslide in 2004. Conversely, does the fact GW received fewer votes than Al Gore indicate that perhaps the American public did not perceive that really big threat a few years in the future? It’s easy to knock down a story that is built on nonsense.

All of which brings us back to the point of this post. Michael Kanell and I have noted that lower tax burdens are not correlated with more rapid economic growth. In fact, from 1929 to the present (and in the book, we focus on the period from 1952 to the present) administrations that have cut the tax burden have performed worse than administrations that raised the tax burden.

And I think we, together and separately, have answered every reasonable objection that has come up, and even quite a few unreasonable objections to boot. And we’ve done so in a consistent and open manner. In our book and in my posts, we’ve been open and clear about our methods and data sources, and we’ve made an effort to treat the data as consistently and systematically as we were able. At some point, the burden of proof should no longer lie with us, but rather on those who cling to a story that simply is not consistent with the data we have observed in the U.S. over the past few decades. Frankly, I think we’re well past that point.

Tags: , , Comments (110) | |