Public Policy and Nichols' Law of Catcher Defense

A post where I disagree slightly with a WCI colleague. There is some hesitation here - not because I am afraid to publicly disagree with a colleague, rather it is because she's smarter than I am and therefore I am not entirely convinced I am correct.

Frances Woolley on boutique tax cuts:

Budget 2011 continues the trend towards an ever more personalized tax system with a new Volunteer Firefighters Tax Credit, a Family Caregiver Tax Credit, and a Children’s Art Tax credit. Coming soon: the Marginal Riding Tax Credit, available to all voters in must-win constituencies...

The importance of the credits is symbolic: they show that smart politics, bad economics outplays bad politics, smart economics every time.

There is an assumption here that these policies are smart politics.  Two reasons for the assumption are given:

But boutique credits are smart politics. First, they appeal to people’s sense that they deserve a break, validate their choices, and reaffirm their sense of self-worth... Second, they give people a sense of control. Tax liabilities stop being something outside of an individual’s control.

It is certainly possible, but I haven't seen a great deal of evidence that these policies are particular popular with the voting public.  I wouldn't suggest that they are unpopular, but are they really vote winners?

Discussions of the popularity of public policies reminds me a great deal of Nicholls' Law of Catcher Defense:

Originally written by Sherri Nichols on rec.sport.baseball sometime prior to 1992, Nichols' Law of Catcher Defense states that a catcher's defensive reputation is inversely proportional to their offensive abilities. Therefore light-hitting catchers get good defensive reputations and top hitters like Ted Simmons wind up with lousy reputations. One can note that many catchers have lost a reputation for glovework once they began hitting better and that the opposite is also true, further reinforcing Nichols' Law.

In baseball, offense is easily measured and observed, where defense is much harder to judge. Under Nichols' Law, people see a trade-off that is not necessarily there and use offense (which is easily evaluated through statistics) to judge defense (which is much more difficult to evaluate).

You have to wonder if the same thing is going on in public policy - that we're using the economic quality of a policy (which is fairly straight-forward to judge) to evaluate political popularity (which is harder to quantify).

If Nichols' Law has parallels in fiscal policy, then economists may be too quick to assume that any political policy that does not make sense on economic grounds must be good politics.

I fully agree with Prof. Woolley when she states:

There’s no point in telling politicians that a particular policy is “smart politics, bad economics.”

There is no point because as economists we may be ceding half the argument, that a particular policy is good politics, when it may not be true.

Comments

[Joseph]

I admit that I have often wondered about the aggregate effect of boutique tax cuts. Any gains in feeding money back to favored groups is offset by the increased complexity and confusion inherent in the tax code. This confusion is especially apt as there can be legitimate deductions that I would never have been able to deduce from the forms or information booklet.

Complexity favors those with top flight accountants and gives the opportunity for political speeches. But the net impact might be negative. Not only is the tax code hard to understand but one might be worried that a deduction isn't legitimate (and audits are widely feared).

[Nick Rowe]

One of these days we should find a topic for a WCI tag team cage match! (Just emailed you Mike).

[MikePMoffatt]

I wonder if there is a topic where we'd have even 2 of us vehemently disagreeing with each other?

I once suggested that Stephen and I do a 'Coyne vs. Wells' type podcast. Problem is it would be 20 minutes of us showing obvious frustration with various members of the media and every political leader in Canada. I don't think it'd make for great viewing.

[Nick Rowe]

Put catching ability on the vertical axis, and throwing ability on the horizontal. Each person is a point. The whole population will look like a shotgun blast. A scattergram with no obvious positive or negative correlation. (Or maybe it's a positive correlation). Now, throw away all the people who don't get picked to play baseball. Remember that the people who pick players have downward-sloping indifference curves, so they will pick the people nearer the North-East edge of the scattergram.

If the population were very large (strictly, a continuum of people), relative to the variance of abilities, a person A who was dominated by another person B on both dimensions would never be picked to play. And the scattergram for players would necessarily exhibit a perfect trade-off between pitching and catching ability. But if the population were finite, we would see some players getting picked (and paid lower salaries) who are dominated in both dimensions by other players (who get higher salaries.

But the selection process will tend to create a negative correlation or trade-off even where there is none in nature.

So the people who act according to Nichol's Law may be rational Bayesians forming beliefs under imperfect information.

My old, related post: http://worthwhile.typepad.com/worthwhile_canadian_initi/2010/06/why-do-economists-assume-tradeoffs.html

[Frances Woolley]

Mike, very much enjoyed this post, and thank you for drawing WCIers attention to that article.

You're right about the lack of hard evidence - and the idea that we simply assume something must be smart politics because it's bad economics is an interesting one.

The 'smart politics' aspect of boutique tax credits doesn't have to be their popularity. Think what the discussion of this budget would have looked like without the boutique tax cuts: people would have been spending much more time talking about the corporate tax cuts and the on-going cuts to the federal public service. So they can be smart politics because they distract people from the real underlying issues, or because they signal the government's future intentions "we really do want to cut personal income taxes, but this is all we can afford to do at the moment."

I suspect the political economy of boutique tax cuts is very much like the political economy of appeasing lobby or special interest groups. The (smaller) group of people who gain care a lot, and make lots of noise about it, the (larger) group of people who lose just kind of ignore the whole thing, because their individual loss is too small to get worked up about, and taxes are boring anyways. The Volunteer Firefighters Tax Credit definitely fits this profile.

I figure I'm on pretty firm ground with the popularity of the Children's Art Tax Credit, too, as middle class parents were annoyed about the unfairness of credits for hockey but not for music, it just didn't feel right to them.

Mike, after my last post, do you think I'm going to let you get away with the "dumb men" defense?

Let's just say the set of things we disagree vehemently about is non-empty....

[R I]

I think we assume that a proposal that is a poor policy must be good politics (or at least the government must think it is) because there is no other way that it would have been proposed. The Finance Department's proposals developed as a result of support from one of two groups: the Department, or the Minister's office. One of those organizations is concerned with good economics, the other with good politics. A proposal which is not good economics likely started with the Minister's office, therefore we would expect that the only reason it received such support is because it was good politics.

I think there is perhaps a third view on the good politics/bad economics point. I left the Tax Policy Branch at the Department of Finance around the same time that Chris Ragan left as visiting economist. He gave us a speech about this notion of bad economics, good politics. I wish I could find an article he wrote on that topic that I could link to here, but his argument was that good politics = good economics, and that good economics, but bad politics, is really bad economics.

[Phil Koop]

I think you are conflating different parties in this analogy, and that this has led you to a specious comparison.

If I am a baseball fan, and I see a team whose catcher is a light hitter, I naturally think there must be some reason the team has put him on the roster. This is a sound inference! That is Nick's point: the guy has beaten many other candidates in a competitive process. Now, if I go on to speculate that the reason is that he is good defensively, that is not a certain inference, it is a conjecture. Perhaps the team has chosen him because he is good at calling for pitches. But the thing is, the team knows why they chose the catcher; they are not assuming he must be good defensively just because he is a bad hitter.

Now apply this to the policy choices made by a politician. As an outsider, if I see that a policy will make for poor government, for economic or any other reasons, I might guess that the politician thinks it is good politics - after all, there must be some reason he chose the policy. But the politician is not guessing! He knows that he thinks its good politics. He might be wrong but so what? There is no inference that he chose the policy because it's bad economics.

[Mike Moffatt]

Hi Phil,

I agree with absolutely almost everything you said, so I'm not sure I'm conflating anything. :)

"But the thing is, the team knows why they chose the catcher; they are not assuming he must be good defensively just because he is a bad hitter."

This is, in fact, *exactly* what the teams and scouts were assuming, particularly in the 1960s and 1970s. Teams believed that Catcher X was a good defensive catcher precisely because he couldn't hit. And vice-versa. Sports writers, scouts and GMs all through the 1960s and 1970s believed Simmons was a lousy catcher, but with advanced defensive metrics that were developed later, we've seen that he was actually average-to-slightly above.

In the analogy, teams=parties, catchers=policies.

The idea here is that we assume politics and economics are orthogonal to each other, so a public policy that is bad economically must necessarily be good politically. I think we need to start to question that assumption.

Thanks for your kind words everyone!

[Joel Wood]

Interesting theory Mike. I think you may be on to something, although that may just be my baseball related bias talking.

[Bob Smith]

I'm curious about the "bad economics" conclusion.

I mean, are tax credits for children's arts programs or sports programs neccesarily worse policies than, say, direct public funding for children's art's programs or sports programs (which have been endemic for decades at the local level) or public funding for sports and art in public schools (which I suspect few of us would critize on policy grounds). Is a tax credit for sports worse than a "fat" tax on fast foods (a program that many economists would probably rationalize, quite reasonably, as a sound policy), or a ban on sugary drinks in public arenas (currently being debated in Toronto), if all are different ways of attacking the same problem (kids are too fat). They may all have diferent impacts (in terms of efficacy, distribution, etc.), but a priori it isn't clear that one is materially wors than the other simply because it uses the tax system.

I take the point that boutique tax cuts can cause complexity in the tax act through, frankly, they're not that complicated and don't really cause that much added complexity. I suppose they do distort spending decisions, but then again, that's the point. In fact, arguably the reason politicians like them (apart from the fact that it gives the federal government a direct link with voters that wouldn't be accomplished by, say, providing funds to local governments for sports and arts) is that using the existing tax mechanism is cheaper, easier, and less complex than any other ways of providing funds to these sorts of programs.

Of course, one can query whether, at the end of the day we should subsidize children's art and fitness programs, or volunteer fire-fighters or caring for the aged (though, I'm not sure I'd want to be on the con side of that last one), but once you accept the merit of those policies, it's not clear why implementing the policies through tax credits are, a priori, worse than using other mechnanisms to achieve teh same goal.

[Phil Koop]

Mike,

I readily concede that your knowledge of baseball vastly outstrips mine. Even so, I am not yet persuaded. For instance, shortstop is a position that was traditionally claimed to put a premium on defense, and thus a position where it might be worth carrying a weak hitter for the sake of his glove. However, if you were to tell me that teams chose shortstops by picking the worst hitter they could find, assuming that on that account he must be the best fielder, I would find that claim extraordinary and I would not accept it without extraordinary evidence. Yet that is the relevant claim here. The distinction is not between fans, writers, scouts, and GMs, but between outsiders and insiders, or observers and actors, whichever way you care to put it. It would be unreasonable for a GM to select a shortstop by scouring the corners of the earth for the worst available hitter, yet it would be quite reasonable for an opposing GM or scout, attempting to assess a player of whom they know nothing except that his a poor hitter, to assume that he is probably a good fielder.

That, in my view, is where the conflation lies: you have conflated baseball outsiders with political insiders. Even if your assertions regarding catchers were proved true, that would not validate the analogy because it would remain to demonstrate that political insiders were similarly liable to the analogous error. That means exactly to demonstrate that a policy is chosen because rather than despite being bad economically. This you have not done; you have not even provided an anecdotal example!

[MikePMoffatt]

Hi Phil,

I'm talking about both baseball insiders and outsiders and political insiders and outsiders. I think you're misunderstanding my argument.

[Frances Woolley]

Bob: "a priori it isn't clear that one is materially wors than the other simply because it uses the tax system"

A lot of the problem with organizing sports is just getting things started. So, for example, an outdoor hockey rink works best if there's boards, and requires some kind of water hook-up and hose for flooding. If that infrastructure can be provided, then community volunteers can do the rest. So there's a good case that public support for infrastructure to get things started is much better than little tax credits that provide support after the fact.

Comparing a fitness tax credit to a fat tax on soft drinks...

The fat tax on soft drinks will create incentives for soft drink drinkers to drink less.

Presumably the goal of a children's fitness tax credit is to induce slacker kids to register for hockey or soccer or other organized sports?

Minor hockey at the house level costs $500 in Ottawa just for registration fees. A complete set of new gear including skates would cost probably at least $300 if you got a good deal, but it's very easy to drop $300 just for a pair of skates or $100 to $200 for a stick. Reasonable quality used equipment would still set you back around $150 to $200 - there's a lot of small items e.g. mouthguards ($10 to $20), throat guards, socks, jocks, that it's not easy to get used. Plus there's team fees and tournament fees and the massive commitment of parental time to drive kids all over the city. If I had to guess the average household income of kids on my son's hockey team this year I'd say $150 to $200K. It could well be higher.

***A lot of the Children's Fitness Tax Credit money is going to relatively well-off families who already have their kids enrolled in sports. For them, it is a pure windfall gain.*** (If you're interested in verifying this, go to the CRA web site, search for Income Tax Statistics, and look at the distribution of credits claimed by income.)

Now having said all of this, when I was involved on the executive of our local hockey association, we didget requests for help with fees - people who had fallen upon hard times one way or another, single parents, etc. We did what we could, which wasn't much, but also referred people to the Max Keeping foundation, and some other organizations that were able to provided limited help. If the money that is being given out in tax credits was instead given to organizations like Max Keeping, it could make a difference. The kind of families we saw who were really struggling probably aren't paying much in taxes anyways - lots of single parents, people who'd lost their jobs, etc - and so a tax credit won't help them.

Continuing on the rant here: being active and organized sports are two entirely different things. The best hockey players are the kids who spend their time mucking around at the outdoor rink. Kids around the world get good at soccer just by kicking a ball around. And then there's Canada's thousands and thousands of lakes and rivers just waiting to be paddled....

[Mike Moffatt]

Perhaps I can explain it in a different way:

Both public policy and baseball defense (particularly prior to the Sabermetrics revolution) have two dimensions which they can be evaluated - one dimension (call it X) can estimated with a reasonable degree of accuracy and a dimension (Y) which has a much larger error band.

Nichols' Law of Catcher Defense states that historically people have tended to allow X to cloud their judgement of Y - that is, for low values of X, E[Y] > Actual[Y] and for high values of Y, Actual[X] > E[X]. It's not that teams went out looking for the worst possible hitting catcher, rather their estimates of the defensive value of a catcher were upwardly or downwardly biased by hitting. So if you had to guys who were an actual 7/10 in defensive ability, the guy who hit .240 would be judged an 8 and the guy who hit .290 would be judged a 6. If a guy could hit *and* was an outstanding defensive catcher, he'd still be seen as a good one - only that he'd be seen as a 9 defensively instead of a 10 (see Bench, Johnny).

I believe the same thing might be going on for public policy - that we overestimate the popularity of policies which a weak economically. Do I have a lot of evidence for this? I wish I did, but I don't. But at the same time, there isn't a lot of evidence to suggest that someone of these policies which are seen as beneficial politically actually cause positive changes in voter behaviour.

[Mike Moffatt]

In the case of Simmons, he was seen around the league as being a particularly weak defensive catcher and he did have some shortcomings - for one thing he had a fairly weak arm. But he was good in other areas - he released the ball quickly, his accuracy was good (not great) and he fielded his position well and he managed to finish in the Top 10 for most defensive categories (putouts, assists, fielding %, caught stealing, etc.) for many years.

[Chris J]

Catching is very difficult and catchers spend lots of time working with the pitchers so they get less batting practice.

But defensive metrics are getting better so we can be pretty sure that Derek pasta-diving Jeter is the best shortstop of the modern era despite being terrible to average at defense. Whereas Ozzie Smith's WAR just isn't as good despite being the best defensive SS of the modern era.

As for catchers, Joe Mauer is the exception which proves the rule.

I think Frances' point should be tempered with diminishing returns. Boutique tax cuts are good politics until the tax code gets so complicated you cannot figure them out.

More baseball analogies please!

[Alan]

As far as I can tell, you are addressing a pretty common phenomenon. Other examples:

-Pretty women are assumed to succeed b/c of their looks.
-Plain women are assumed to be competent due to success despite their looks.
-Cheap things are assumed to be low quality.
-Fast cars are bad on gas.

My favourite: Always get the freaky looking (punky, tattooed, or overly-pierced) guy in the IT department to help you - they must be competent to succeed despite their appearance and presumed attitude.

Like all generalizations, they are not always correct. Some may even be always wrong.

[Nick Rowe]

I think Alan is right. Nichols' Law should be generalised. It's a Law of Economics. And this one is my favourite too, because it explains the reasoning behind the inference:

"Always get the freaky looking (punky, tattooed, or overly-pierced) guy in the IT department to help you - they must be competent to succeed despite their appearance and presumed attitude."

Gotta be a paper in this!

[Yvan St-Pierre]

I find this discussion quite interesting. However, it seems to me that the analogy has two important limitations.

First, in baseball, offensive and defensive abilities are joint inputs of success. In policy-making, political quality is the measure of success itself. Whether it's good in any other dimensions may indeed contribute to that success, but if it does so even slightly at the expense of being good politics, then it just doesn't contribute to anything else than failure. In other words, when you agree that a bad economic policy is smart politics, you're not ceding half the argument, you're ceding all of it.

Second, if we take an outsider's perspective, so as to have 2 really independent dimensions, I would think that what is more easily measurable here is whether or not a given set of policies are politically good or bad. If they are, the party gets more seats and wins, if it isn't, you don't. Then, whether it's good from an economics point of view is generally a lot harder to assess, if only because economists have a reputation for having two hands, right? Or maybe I'm really missing the point because I'm too excited by the upcoming election... Now, I get that the point is about the effect of a specific policy, rather than a set, but the identification problem here is equally present in both dimensions (or else there would probably be a lot less bad economic policy in the first place).

[marcel]

2 points:

1) Nick Rowe wrote: Always get the freaky looking (punky, tattooed, or overly-pierced) guy in the IT department to help you - they must be competent to succeed despite their appearance and presumed attitude."

Isn't this why (at least in the US), blacks and women have long thought that they had to be twice as good (as a white male) to be considered as good?

2) Mike Moffat wrote:

A post where I disagree slightly with a WCI colleague. There is some hesitation here - not because I am afraid to publicly disagree with a colleague, rather it is because she's smarter than I am and therefore I am not entirely convinced I am correct.

Is this MM doing a first, WCI dumb man post?