Old post by Bill Rick check this out.

from Dick Robertson,2009.10.02.0911CDT]

[From Rick Marken (2009.10.01.1540)]

My data on the relationship between taxation and growth/unemployment showed that there was no statistically significant relationship between taxes and these variables. All economists say with great authority that increasing taxes reduces growth and employment. My data show that they could not possibly be making this claim on the basis of what is observed. Republicans have happily used this purported economic truth as the basis for a tax reduction policy that has driven the country into bankruptcy. My data simply shows that the economists should just be honest and say “I have no idea what effect taxes have on the economy”. Nothing Richard said about correlations obviates this point.

Interestingly (to me, anyhow) there was an article by Robert J Barro, and Charles Redlick on page A23 of yesterday’s Wall St. J. titled Stimulus Spending Doesn’t Work, the last sentence of which was, “However, there is empirical support for the proposition that tax rate reductions will increase real GDP.” The WSJ identifies Mr. Barro as “a professor of economics at Harvard. Mr Redlick is a redent Haravard graduate.”

What do you think?

Best,

Dick R

[From Rick Marken (2009.10.02.1615)]

Dick Robertson (2009.10.02.0911CDT) --

Interestingly (to me, anyhow) there was an article by Robert J Barro, and
Charles Redlick on page A23 of yesterday's Wall St. J. titled Stimulus
Spending Doesn't Work, the last sentence of which was, "However, there is
empirical support for the proposition that tax rate reductions will increase
real GDP." The WSJ identifies Mr. Barro as "a professor of economics at
Harvard. Mr Redlick is a redent Haravard graduate."

What do you think?

This is very interesting to me too. Thanks, Dick.

I would love to see the "empirical support". I will write and ask them for it.

I see one possible problem. They say there was evidence that " tax
rate reductions will increase real GDP". I was looking at the
relationship between tax rate and _rate of change_ in GDP. But still,
they should have found the same results as I did; if I did mine in
terms of raw GDP I would have found a positive relationship between
tax rate and GDP (rather than the negative relationship they say that
they found); I'll check it out when I get back to my computer with the
data.

It's possible that they did find a negative relationship between tax
rate and GDP (lower taxes associated with larger GDP). That finding
would be really interesting; my guess is that if that were the finding
it would be because they used a different measure of tax rate. I used
the tax rate on the upper bracket, which is kind of a "progressive"
measure of tax rate. Perhaps they used something like a "base" tax
rate. This is the tax that would affect low income people the most and
I could imagine that increasing that rate would be recessionary.

But we'll see. I'll write to them now.

Oh, and those of you think this data is just noise can put your
fingers in your ears and say "la la la la la" when I report it;-)

Best

Rick

PS. I will continue with my economic story soon; stay tuned Shannon!

···

--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

[From Rick Marken (2009.10.02.2000)]

I wrote to Barro asking for the evidence that "tax rate reductions
will increase real GDP". I hope to hear from him soon.

But in the meantime I took a quick look at the correlation between tax
rate (upper bracket) and GDP (annual from 1948 to 2008) and found that
the correlation is -.85! That's a correlation that even Bill Powers
might like (a little;-). It shows clearly that there is a strong
negative relationship between tax rate and GDP, just as Dr. Barro says
(although Barro implies that the relationship is causal -- that
reduced tax rates _will_ increase GDP-- when, of course, there is no
such evidence).

I realized that the increase in GDP could just be a result of the
increase in population that occurred over the same period while the
decreased tax rate is a result of policy changes during that period.
So I computed the partial correlation between tax rate and GDP
factoring out population. The result is that the correlation between
tax rate and GDP goes almost to zero (-.076). So when you factor out
population there is not even evidence of a relationship between tax
rate and GDP, let alone a negative causal one.

Recall that the correlation between tax rate and dGDP/dt (growth) was
.28 for the same period (1948-2008). When I factor out population
growth from this correlation the result is a correlation that is
somewhat higher, .36. These correlations are still small and I am
completely puzzled by the difference in the correlation of tax rate
with GDP vs GDP growth.

But this exercise was useful to me. First, it shows me where
economists get the idea that increasing taxes is recessionary. It must
be based on that huge, artifactual correlation between tax rate and
GDP. Apparently, the economists don't want to deal with the
correlation between tax rate and GDP _growth_. Second, the fact that I
see all these puzzling relationships convinces me that, once again,
Bill Powers is right; there is no way to make sense of the data
without a model. This is what I believe as well, of course. But I
guess I thought there might be more value to the data than there
actually seems to be. So the only thing I think I'm right about is
that economists are completely clueless about how an economy works.
And that includes Krugman, though I think he has a better idea of how
clueless economists are than do most economists. So I don't think we
control theorists have to be humbled in any way when we poke our noses
into the temple of Economics.

Anyway, I guess I can stop playing with the data for a while and get
back to the modeling.

Best

Rick

···

--
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

[From Bill Powers (2009.10.03.0717 MDT)]

Rick Marken (2009.10.02.2000) --

I wrote to Barro asking for the evidence that "tax rate reductions
will increase real GDP". I hope to hear from him soon.

But in the meantime I took a quick look at the correlation between tax
rate (upper bracket) and GDP (annual from 1948 to 2008) and found that
the correlation is -.85! That's a correlation that even Bill Powers
might like (a little;-).

At that correlation, the odds are a shade less than 5 to 1 that a decrease in upper-bracket tax rate (of any size) will go with some amount of increase in GDP rather than a decrease.

Is there any way of finding a record of how much taxes the largest earners actually paid? That is, not the tax rate from the tax tables, but the actual percentage of income paid as taxes by those in the highest tax brackets. When you can hire the best tax advisors you don't really have to pay the highest rates. That's why you have congressmen on your payroll, too, to create the loopholes.

You're getting pretty good at finding the data and analyzing it!

I realized that the increase in GDP could just be a result of the
increase in population that occurred over the same period while the
decreased tax rate is a result of policy changes during that period.
So I computed the partial correlation between tax rate and GDP
factoring out population. The result is that the correlation between
tax rate and GDP goes almost to zero (-.076). So when you factor out
population there is not even evidence of a relationship between tax
rate and GDP, let alone a negative causal one.

I think we need an additional piece of data here, though I could be revealing my ignorance about statistics. The correlation shows how much scatter there is in the data, but the regression line shows what the slope of the assumed linear relationship is, as I understand it. You might have a regression line with a slope of -1.0 going with a correlation of -0.076, mightn't you? This would seem to tell you that one variable has a relatively large effect on the other, but the low correlation says that this is an almost totally unreliable conclusion. I'm thinking of Richard Kennaway's ellipses, in which the long axis is tilted to indicate the regression line, and the ratio of major axis to minor axis, or one minus that ratio, indicates the correlation (both indications being approximations). A correlation of -0.075 tells you that the scatter plot is almost circular, but doesn't tell you what the slope of the major axis is -- or does it?

What are the slopes of the regression lines?

Recall that the correlation between tax rate and dGDP/dt (growth) was
.28 for the same period (1948-2008). When I factor out population
growth from this correlation the result is a correlation that is
somewhat higher, .36. These correlations are still small and I am
completely puzzled by the difference in the correlation of tax rate
with GDP vs GDP growth.

With that correlation, the odds that an increase in d(GDP)/dt will go with an increase in tax rate rather than a decrease are only about 1.5 or 2 to 1, meaning that if you make many predictions of the rate of change of GDP from observations of tax rate, you will be right only 60% to 67% of the time. Might as well toss a coin if you have only one chance, especially since you're only predicting the sign, not the magnitude of the effect. Your ten-dollar investment might earn you only 10 cents even if you're right.

But this exercise was useful to me. First, it shows me where
economists get the idea that increasing taxes is recessionary. It must
be based on that huge, artifactual correlation between tax rate and
GDP. Apparently, the economists don't want to deal with the
correlation between tax rate and GDP _growth_. Second, the fact that I
see all these puzzling relationships convinces me that, once again,
Bill Powers is right; there is no way to make sense of the data
without a model.

Thanks for that. Your results tell us that growth in GDP is due to something other than taxes. Common sense models tell us that it's certainly not due to population growth; having a baby does not increase GDP unless something else happens at the same time to give the parents more income to spend for the next 18 or 20 years. Even a delayed correlation doesn't help. When the baby gets a job and starts earning money to spend, where does that extra income come from? So there is some as-yet-unknown factor that makes GDP increase with time, and perhaps changes the tax rate, too. I know what it is! It's the change in the distance to the moon (it increases by about 38 millimeters per year). That must be a very powerful influence, since such a small change alters GDP by billions of dollars per year. Astrologers were right!

Statistics can't teach us anything, though it might tell us that we don't know something. The universe remains totally mysterious when seen only through statistics -- everything happens because it happens. Human beings are to chimpanzees as modeling is to statistics. Both have been evolving for the same length of time, so they are both well-developed examples of their species. But one has a lot more useful abilities than the other has. Of course in some areas each has advantages; for example, if a human being wanted to eat termites, chimpanzees could teach him or her how to use a stick to get them out of the mound.

Best,

Bill P.

[From Rick Marken (2009.10.03.2245)]

Bill Powers (2009.10.03.0717 MDT)]

Rick Marken (2009.10.02.2000) –

I wrote to Barro asking for the evidence that “tax rate reductions
will increase real GDP”. I hope to hear from him soon.

But in the meantime I took a quick look at the correlation between tax
rate (upper bracket) and GDP (annual from 1948 to 2008) and found that
the correlation is -.85! That’s a correlation that even Bill Powers

might like (a little;-).

At that correlation, the odds are a shade less than 5 to 1 that a decrease
in upper-bracket tax rate (of any size) will go with some amount of increase
in GDP rather than a decrease.

I got the Barro paper and, as in wonderland, things just get curiouser and curiouser. It turns out economists do the same kind of crazy statistics I do. Barro uses the good old general linear model in the paper that provides the evidence that “tax rate reductions will increase real GDP”.

The first discovery I made is that Barro is predicting GDP growth (as I do), not GDP.His research is based on the following linear model:

(yt – yt-1)/yt-1 = â0 + â1·(gt – gt-1)/yt-1 + â2·(ôt-1 – ôt-2) + other variables,

where (yt – yt-1)/yt-1 is GDP growth(or decline)/ year, (gt – gt-1)/yt-1 is increased (or decreased government spending/year and (ôt-1 – ôt-2) is an increase or decrease in Barro’s measure of marginal tax rate. Barro’s marginal tax rate measure is a rather complex combination of federal rates, state rates and withholding. The article contains these marginal tax rate measures for each year from 1916-2006. So I was able to compare them with the tax rate measure I was using (top marginal rate). The correlation between Barro’s tax rate and mine is -.31; so as the top marginal rate went down (which it generally has since 1934) the Barro tax rate went up; assuming that the Barro rate is close to the “average” tax rate, the negative relationship suggests that tax rates have become more regressive. , which seems to correspond to what’s been going on.

I found (Barro did not report it) that the raw correlation between the Barro tax rate and GDP growth rate is .21 – positive, rather than negative. Recall that the correlation between my top tax rate and GDP growth was .3. So in both cases the relationship is slightly positive (increased taxes associated with increased growth). The raw correlation between change in Barro tax rate and GDP growth is still positive: .49. The correlation between change in top tax rate and GDP growth is also still positive: .14.

So where does Barro get the idea that there is a negative relationship between tax rate and growth? It comes from the multiple regression analysis using the linear model above, which includes government spending as a predictor. And he only finds a statistically significant negative relationship (indicated by a significantly negative â1) in one case – when he looks only at data from 1950 to 2006. So the “evidence” for the negative effect of taxes on growth comes from running a rather complex analysis; the raw positive relationship between tax rate and growth apparently is “factored out” by the multiple regression due to the way government spending is correlated with tax rate and growth. I did some multiple regression analyses myself without using government spending as one of the variables and in none of them was I able to come up with a solution where the coefficient for (correlation between) any of the tax rate measures or change in tax rate measures was negative when predicting GDP growth.

Is there any way of finding a record of how much taxes the largest earners
actually paid?

Yes, that might be useful data to get. I’ll try.

You’re getting pretty good at finding the data and analyzing it!

I’m OK at finding it. But I can see from this experience with the Barro data that (to paraphrase Roy Schieder in “Jaws”) we need a bigger model.

A correlation of -0.075 tells you that the scatter plot is almost circular,

but doesn’t tell you what the slope of the major axis is – or does it?

You are exactly right. Actually, I should take a look at some of the scatters to see what the data actually look like; the correlations might be lower than they should be if there are nice non-linear relationships between the variables. Population growth over time, for example, is definitely non-linear.

Statistics can’t teach us anything, though it might tell us that we don’t
know something.

The statistics I am using are based on a model (the open loop causal model shown above in the form of the general linear model). It’s not the statistics per se that are the problem; it’s that we are using the wrong model.

Best

Rick

···


Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

[From Bill Powers (2009.10.04.0255 MDT)]

Rick Marken (2009.10.03.2245) --

I got the Barro paper and, as in wonderland, things just get curiouser and curiouser. It turns out economists do the same kind of crazy statistics I do. Barro uses the good old general linear model in the paper that provides the evidence that "tax rate reductions will increase real GDP".

Well, it's good to know that the other side uses the same kind of data and calculations you use. This puts you on even footing with them.

I found (Barro did not report it) that the raw correlation between the Barro tax rate and GDP growth rate is .21 -- positive, rather than negative.

At that correlation, the odds that an increase in tax rate will go with an increase in growth rate are 1.3:1. The chances of correctly predicting the sign of the relationship between a tax increase and the growth rate are therefor 1.3/2.3 or 56.5%, and the chance of predicting incorrectly are 43.5%.

Recall that the correlation between my top tax rate and GDP growth was .3.

Chance of prediction of sign of effect: right, 60%; wrong, 40%.

> Statistics can't teach us anything, though it might tell us that we don't
> know something.

The statistics I am using are based on a model (the open loop causal model shown above in the form of the general linear model). It's not the statistics per se that are the problem; it's that we are using the wrong model.

The statistics are a problem in that they can't tell you something is wrong with your model. They will work just as enthusiastically for a bad model as a good one.

They are also a problem when the correlations are low, because they tempt you into drawing conclusions from relationships that just barely exist at all and hold true only slightly more often than they don't. At a correlation 0.14, your chances of predicting correctly are only 55%, no better than flipping a coin. That's not a positive correlation, it's NO RELATIONSHIP.

And whatever happened to "correlation is not causation?" Maybe an increase in GDP sometimes encourages politicians to raise tax rates, since the people are more prosperous. Of course it could also encourage politicians to lower tax rates since revenue is up. Whatever happens, we can explain it (after it happens).

Best,

Bill P.