Tuesday, December 19, 2006

Hoosier Follies

By Richard Vedder

When he was with the Bush Administration, Mitch Daniels had a reputation for being a hard money man, a guy who would occasionally say no to some spending initiatives amidst a group of Big Government Conservatives whose unprincipled spending behavior contributed to the GOP's debacle last month.

Now as governor of Indiana, Daniels is threatening to repeat the sins he fought in Washington several years ago. To get Indiana moving ahead economically, he wants to spend a bunch of money on higher education. By privatizing the state lottery (probably a good idea), he wants to take a billion dollar windfall and drop it into higher ed, partly to fund HOPE-style scholarships and to lure big name scholars to the Hoosier State.

The intentions are good, and the goal of improving Indiana's so-so economic performance is laudable. However, the proposal, if adopted, suffers from several flaws.

1) National data suggest that the relationship between spending more state funds on higher education and economic growth is either non-existent or is negative -- more spending means lower growth;

2)An increase in scholarship aid will likely be at least partially offset by larger tuition increases than otherwise would be the case, as universities try to capture some of the funds that Governor Daniels wants to spend;

3)HOPE type scholarship programs extenuate the move away from need-based financial aid and tend to make flagship universities (e.g., Indiana University at Bloomington (U) and Purdue at Lafayette) into more bastions for upper class kids who tend to be the major recipients of merit-based scholarships. Should Indiana be selling state assets to provide assistance to children of yuppies attending increasingly country club-like institutions? Is this the optimal use of state resources?

4) Targeted programs to bring eminent scholars into states seem to have no effect on economic growth. For example, neighboring Ohio's Eminent Scholar program of a decade or two ago was implemented --and income per capita relative to the nation continued to fall. It is not clear to me how Indiana benefits if Professor X does his or her work at IU rather than at the University of Illinois, unless there is a huge royalty payoff from patents, which seldom is the case in the short or intermediate run.

I suspect that if Governor Daniels really wanted to increase Indiana's competitiveness, he would be better served reducing the tax burden, which historically once was low in Indiana but is no longer so.

13 comments:

Bob Yates said...

I sure wish I understood how economists use data for their arguments, especially when they present different arguments in different venues. Our gracious host tells us:

National data suggest that therelationship between spending more state funds on higher education and economic growth is either non-existent or is negative -- more spending means lower growth;

I sure wish at some point we could learn exactly what the actual correlation co-efficient is and how many states don't show this negative correlation, but in the meantime let's consider some actual data about funding for higher education in this new century.

In late October Inside Higher Education had an article about how the recession has cut back on state support for higher education.

Here are some numbers about funding for some other states. The states I cite are useful for another document on the web our gracious host has just posted.

The first percentage is change in state support in 1980, the year is when funding reached 1980 numbers; the second percentage is change in funding in 1991-2, the year it was returned to that level; the final percentage is change in 2001-3, if there is a year when funding returned to 2001-3 level.

MO -15.5% 1986 -3.5% 1994 -23.4% *

Tenn. -3.7% 1984 -1.5% 1994 -0.6% *
Okla. +19.9% * +15.5% * -18.8% *

Wyo. +25.3% * -5.4% 2001 +12.9%

If what our host has written here is correct, Missouri should be better off than any of the other states and Wyoming should have terrible growth numbers and Tennessee should follow Wyoming.

However, in that other document written by host (available here http://showmeinstitute.org/smi_briefing_4.pdf)
that is not the case.

In that other document, our host proposes to end the income tax in Missouri. He notes that Oklahoma and Tennessee, neighboring states without an income tax, have growth rates better than Missouri's. However, they appear to have increased funding for higher education at higher rates and that, according to his post here, should retard such growth. Yet, both grew faster than Missouri from 1995-2005: Missouri 23%; Oklahoma 37 (a possible missprint; Tennessee 32%.

I put in Wyoming, because it had a 45% growth rate over the same period. For Wyoming according to the growth data our host uses, the following is wrong: the relationship between spending more state funds on higher education and economic growth is either non-existent or is negative -- more spending means lower growth.

(I wonder if he will share with us in how many other states the relationship between higher funding for higher education and a negative growth rate does not exist.)

Of course, I am not a distinguished economist, but I think the relationship between state growth and funding for higher education is a little more complicated than suggested here and in another venue even our host suggests other reasons for economic growth that have nothing to do with funding for higher education.

Anonymous said...

It would be interesting to see what "national data" corroborate the first "flaw" in the idea of increasing state funding for higher education. The very fact that Gov. Daniels has decided to increase funding in an effort to increase growth suggests that what may be observed in the national data could easily be spurious -- that is to say, those states that exhibit lower growth are more likely to look to ways to increase growth, e.g. by investing in higher ed. I would very much like to know how the problem of endogeneity is controlled for in the data, if at all.

Jim said...

YATES:

"Of course, I am not a distinguished economist." Therein lies your problem. Your analysis is both specious and superficial.

Jim said...

HAHN:

"endogeneity" - Very impressive. I know I'm impressed. You impressed Rich?

How about this one?

supercalifragilisticexpealidotious

Bob Yates said...

Jim, I assume that you know how "distinguished" economists argue, so I take your response, with its insightful analysis of my arguments, as an example of how "distinguished" economists respond to another's argument.

Thank you for spending so much time on why I am wrong. Jim wrote:

"Of course, I am not a distinguished economist." Therein lies your problem. Your analysis is both specious and superficial.

Jim said...

YATES:

Here is what I said - please read carefully: "'Of course, I am not a distinguished economist.' Therein lies your problem. Your analysis is both specious and superficial".

Please tell me where I said you're wrong. Your analysis is what I said it is. But if you want me to conjugate a few more irregular verbs and throw in some adjectives, why, here you go:

Vedder is distinguished for a reason - no quotation marks needed. Please educate yourself with regard to his credentials.

"Thank you for spending so much time on why I am wrong." Well Yates, here is a longer version of what I think of your analysis:

Your analysis is stupid and shallow. It lacks clear thinking and perspective, and it is a simple-minded argument. Your analysis isn't necessarily wrong, it's just plain irrelevant. Maybe you should read Vedder's book "GOING BROKE BY DEGREE" so you don't have to take a knife to a gunfight.

All four points that Vedder makes in this epistle are backed by factual data. Your numbers, on the other hand, are backed by equivocation of data relationships and what the data means. And your logic is twisted to create a self-fulfilling prophecy.

You use your twisted logic to dismiss Vedder's points, yet you even say, "...our host suggests other reasons for economic growth that have nothing to do with funding for higher education." You contradict yourself.

I like to get to the point, and if I think someone is full of shit, I don't need to write a novel to let them know they are full of shit. Hope that helps.

Big Blue said...

Endocrinology...er...endogeneity. Whoah dude, I'll bet that guy is smart! Count me in as VERY impressed!

Bob Yates said...

Jim, Thank you for putting a little more time into your last response.

As an expert on our host's book, would you please report the r-value for the correlation between state funding for higher education and state growth rate? Notice this negative relationship does not obtain for Wyoming.

I understand that when actual data are cited and those data disagree with one's own ideas, there is a temptation to say the other's data are bogus. It is so much easy than actually acknowledging one's ideas might not account for some data.

I guess "factual data" and relationships are in the eye of the beholder. Jim writes the following about my analysis:

All four points that Vedder makes in this epistle are backed by factual data. Your numbers, on the other hand, are backed by equivocation of data relationships and what the data means.

Here is our host on his first point:

National data suggest that the relationship between spending more state funds on higher education and economic growth is either non-existent or is negative -- more spending means lower growth;

I guess when our host writes "suggest" that is not an equivication for our good friend, Jim, and providing no real numbers is "factual." Not being an economist, I have no idea whether Jim's statements are representative of how economists disagree with each other. Is it, Jim?

Finally, again this is from a non-economic perspective, but I learned that correlations -- more funding for higher education correlates with lower state growth -- are not causal relationships. See the above from our host: "more spending means lower growth."

Apparently, and Jim please let me know, for economists, especially our "distinguished" host, this principle does not obtain. A correlation, apparently for our host, means a causal relationship. It is always interesting to learn that what one discipline assumes about statistics is suspended in another discipline.

Jim said...

YATES:

"Here is our host on his first point:

National data suggest that the relationship between spending more state funds on higher education and economic growth is either non-existent or is negative -- more spending means lower growth."

In this regard, one has to use some common sense, rationale, and reason. Once again, your analysis dismisses the point Vedder is making; and your analysis is like saying that if I drink one beer and it makes me feel 10 times better, if I drink 10 beers, I'll feel 100 times better, and if I drink 100 beers, I'll feel 1000 times better. But this is obviously not what Vedder is saying. But it is a standard you are mistakenly holding him to. So common sense, reason, and rationale tell us that your analysis is way off the mark. So why do you hold Vedder to such a standard with no room for reason? If the DJIA is increasing and reaching record highs; would you agree that this data suggests that the economy in general is doing well? Does one down week mean the economy is tanking? What’s the correlation? What’s the r-factor? Is a good economy driving the market or is the market driving a good economy? What about the Fed’s monetary policy? What’s the effect? What are the effects of corporate earnings on the economy, and the market? Now take all of this and try to find your dear-to-the-heart correlation; do a statistical analysis of such a dynamic mess. I on the other hand observe trends based on factual data and draw conclusions – no statistical analysis required – let the grunts in the boiler rooms do it. Vedder has seen the statistics and has cited them in previous blogs – no need to keep regurgitating the same info over and over again. Statistical analysis is a very important tool in economics, but an economist needs to have a larger toolbox and we should not place limits on an economist and say “You must statistically prove your case for it to be valid”. I think maybe Vedder sees something that doesn’t make sense and is asking a good question: “Why?” And I am happy he is doing what he does. The guy has great big balls – hard as tool steel. He is critiquing the very system he is a part of. He’s looking out for the people that have to borrow and put their homes at risk (if that’s enough) to send their kids to college.

"I sure wish at some point we could learn exactly what the actual correlation co-efficient is..." I do not understand why you are so fixated on establishing a number between +1 and -1 and cross validation when you know as well as I do that you are chasing a futile goal. Why can't you live with trends? If you self directed investments, you would have a heart attack.

"...would you please report the r-value for the correlation between state funding for higher education and state growth rate?" Answer: NO.

“I guess when our host writes "suggest" that is not an equivication (sic) for our good friend, Jim, and providing no real numbers is "factual." First, I am neither your friend nor your good friend. I am not your foe either. You don’t know me and I don’t know you. When analyzing trends and/or statistical data, it is quite common to use the word “suggest” or “suggests”. And then he follows with an unequivocal statement; “more spending means lower growth” – no fuzz on that peach. This has everything to do with analyzing trends and data and nothing to do with equivocation. Your use of the data that you provided maybe factual, I don’t know. But your analysis of the relationship of the data was, to me, equivocation as you were trying to use it for comparative purposes – and it doesn’t flush (back to the beer example).

“..but I learned that correlations -- more funding for higher education correlates with lower state growth -- are not causal relationships.” If you limit causal inference to only two absolute variables (higher ed funding and state growth) I would tend to agree with you. To consider other inherent additional variables would force you to throw “correlation” out the window – or you could set up a credible DoE and ask the Feds if you can use their supercomputers. But if you are more than 10 years old, you won’t get your solution in this lifetime.

“It is always interesting to learn that what one discipline assumes about statistics is suspended in another discipline.” Once again with clarity.

Bob Yates said...

Jim, This certainly is fun. All the handwaving in the world, and Jim I really do admire your handwaving, is not sufficient to justify what our gracious host is saying.

Jim writes:

Vedder has seen the statistics and has cited them in previous blogs – no need to keep regurgitating the same info over and over again.

On this blog, Jim, he keeps repeating the claim there is a negative relationship between state support for higher eduction and state growth. However, our host has never said what the correlation co-efficient is, and as best as I can tell, you don't know.

However, I really appreciate your recognition that our gracious host seems to overstate this relationship. Remember our host has said: "more spending means lower growth."

Our good friend Jim (and I do consider you a friend because even though you have called my arguments names you have never attacked me personally) observes that our host's simple relationship is wrong. Jim, correctly citing my observation, writes:

“..but I learned that correlations -- more funding for higher education correlates with lower state growth -- are not causal relationships.” If you limit causal inference to only two absolute variables (higher ed funding and state growth) I would tend to agree with you.

Thank you for recognizing that our gracious host's statement on the negative relationship between state funding for higher education and state growth is flawed. See what happens when two reasonable people have a civil discussion.

Jim said...

YATES:

Merry Christmas & Happy New Year

Jim

Bob Yates said...

Jim,

Let me wish you the same for whatever holidays you celebrate at this time of the year!

Jim said...

Political Correctness turns my stomach.