The Trickiest Tripwires in Analytics

Posted on by David Wright

1. Are my conclusions BS?

I admire this. The author found an interesting statistical anomaly and got some attention for it. Then she discovered it was actually all due to an error and published a retraction of the use of Benford’s law as a fraud detector.

2. How do I figure out causation?

Does teen pregnancy cause poverty or vice versa? Here’s Tim Taylor:

In an ideal experiment, one might want a research design in which a random sample of teenagers becomes pregnant and gives birth, and then you could track the outcomes. Of course, randomized pregnancy is an impractical research design! But here are four approaches used by clever economists to disentangle this question of cause and effect.

A within-family approach. Look at life outcomes for sisters who give birth at different ages. The result of this kind of study is “once background characteristics are controlled for, the differences are quite modest. Furthermore, even these modest differences likely overstate the costs of teen childbearing, since the sister who gives birth as a teen is likely to be “negatively” selected compared
to her sister who does not.”

Miscarriages.  Of those teens who become pregnant, some will suffer miscarriages. Compare women who are similar in measured characteristics of family background, but some of whom gave birth as teenagers while others had a miscarriage. It turns out that their life outcomes look quite similar: that is, giving birth as a teenager doesn’t appear to cause any additional decline in later life outcomes.

Age at first menstruation. Girls who menstruate earlier are at greater risk of becoming pregnant as teenagers. One can use a statistical approach to look at two groups of women who are similar in measured characteristics of family background, but where one group has a higher pregnancy rate because they began their menstrual cycle earlier. However, the life outcomes for these groups look quite similar; is not correlated with lower life outcomes: that is, a random chance of being more likely to give birth as a teenager (because of an earlier age of first menstruation) doesn’t appear to cause any additional decline in later life outcomes.

Propensity scores. Look at girls within a certain school, so that they live in more-or-less the same neighborhood. Using the available data, develop a “propensity score” that measures how likely a girl is to give birth as a teenager. Then compare the life outcomes for girls with similar propensity scores, some of whom gave birth and some of whom did not. There doesn’t seem to be a difference in life outcomes, again suggesting that giving birth as a teenager doesn’t much alter other life outcomes.

Today In Bailout-ology

Posted on by David Wright

A bailout, in principle, isn’t really THAT horrifying, really. It’s the way bailouts get done that is irritating. And in 2008, financial firms’ bondholders got 100 cents on the dollar to keep everyone out of the bankruptcy courts. That’s unpleasant.

Multiple, simultaneous bankruptcy, the story goes, would have been immensely disruptive; that’s what’s meant by the whole “bring the system down” and “destroy the financial system” and “financial armageddon” thing. I don’t know if the real driving force behind this fear (complexity of these firms’ interconnectedness) has been addressed, but I  know which way I’d bet.

Anyway, the 100% bailout for bondholders of financial firms stands in pale comparison to GM and Chrysler’s (secured) lenders, who apparently took a 71% haircut. Here’s more:

It seems clear that the federal government shouldered out bondholders, who would have received more in a standard bankruptcy procedure, and thus created some uncertainty about how bondholders of other large firms might be treated in the future. On the other side, the UAW retirement funds did much better out of the stage-managed bankruptcy than they probably would have done in a standard bankruptcy. Fiat appears to have gotten a better deal under the stage-managed bankruptcy of Chrysler than it would have received in a standard bankruptcy. The stage-managed bankruptcy did lead to cost-cutting measures like plant closures, fewer employees, and more competitive wages for GM and Chrysler, but presumably these changes would have happened under a standard bankruptcy procedure, too–and perhaps they would have happened in a way that led to greater competitiveness for the firm moving forward.

As a general principle, Uncle Sam is happy to dance around legal precedent (and legal laws?) and screw over bondholders. And there isn’t anything special about individual banks’ bankruptcy, either.

The problem, then, is the interconnectedness of financial firms. No politician has the stomach to let them all get wiped out at once. Nor the stomach to prevent such a crisis in the future.

Where The Skills Are

Posted on by David Wright

Here’s a graph to think about:

My goodness!

It would be really interesting to see this broken down by industry, which I’ve had a quick look into but have been stymied by the BLS website for now. I’m sure the data is there but blogging is taking a firm back seat to studying until I’m past my exam later this week.

Baby boomers aren’t going to retire like their parents did because they can’t. And that’s very good for the economy because they’re a damn productive bunch for the most part.

I’ve forgotten where I saw this graph first as it’s been sitting in draft limbo for a few days…

Low Hanging Mirages

Posted on by David Wright

Consider what the effects of this kind of regulation are:

Antonios Avgerinos, 59, a retired army pharmacist, always wanted his own pharmacy here. And why not? Greek law ensures that pharmacists get a 35 percent profit on all drugs sold, even over-the-counter medications.

But Greek law also limits just about everything else about pharmacies. They must be at least 820 feet apart and have a likely market of no fewer than 1,500 residents. To break into the business, an aspiring pharmacist generally has to buy a license from a retiring one. That often costs upward of $400,000.

The goverment, through its extensive regulation of pharmacies, has created an asset that is traded between retiring pharmacists and aspiring ones. There is no good reason why this needs to be the case and this is a fantastic example of free lunch deregulation, right?

Well, as soon as you deregulate these pharmacies, you destroy $400,000 of paper wealth for every pharmacist in the country. The macroeconomics of that are probably tolerable, but the politics are absolutely toxic. No politician wants to tangle with a highly educated group of cornered wolverines fighting for their nest eggs.

Among other strategies, they can probably point to dozens of other equally ridiculous regulations that should be struck down first. Until they wise up and realize that by teaming up with these other interest groups they can create an invinciple coalition against reform.

And what if all of these incumbent interests borrowed money to pay for their licenses? That means that reform will also wipe out the debt they owe, which means the banks lose out. Those banks can hardly afford that, can they! Suddenly fixing a ridiculous regulation has morphed into economic self-annihilation.

Now consider how this story plays out in other countries. Here in the US a lot of wealth has been chenneled into rising house prices, probably for stagnationist reasons, and now you have a similar situation for housing reform. Can’t tell everyone you’re going to destroy 10-15% of their GROSS wealth, particularly if that equals 115% of their net wealth. And then there’s the banks.

I posted on the politics of the anti-loose-money coalition, basically saying that it’s not demographics because the policy responses to 2008 and 1931 were the same while the demographics were far far different.

Well, Steve Waldman was gracious enough to comment, mostly citing cultural and intellecutal advances since the 30s. I’m skeptical and responded by pointing out one huge similarity between the monetary mistakes of today and the 1930s: DEBT.

That graph isn’t an awesome one because in a big bad crisis both the numerator and denominator are changing in erratic ways, relative to history. And deflation really complicates the analysis. I’m not gonig to parse the data myself today, though, so I’ll go on and just make my point.

Which is that too much debt is toxic for lots of reasons, most particularly because it chains policymakers to the status quo. You can’t loosen up monetary policy because you destroy too much wealth and perhaps spur a banking crisis.

You can’t raise rates either: consider China, where interest rates are probably too low. Their growth model has been a debt-fueled investment boom (see Michael Pettis for more on this) which, Japan has taught us, usually ends with too much debt and too much useless crap. But if you have too much debt, raising interest rates is suicide at roll-over.

I’m trying to tie together a common story through all this, which is that there are often problems in an economy and sometimes these problems get really bad. Debt makes it really really difficult to get out of these problems and once the market realizes this, NGDP forecasts drop. Then you have a Krugman-Sumner recession, which I think of as simply an expression of our straightjacket.

If we had control of the economy, we’d get out, but if we had control we’d never have gotten in in the first place.