Megan McArdle dialogue at AEI

You will find the video here, the dialogue is with me and Tim Carney.

The underlying question behind the book is when the general possibility of error helps you learn and when it leads you off the precipice, never to return.  As I understand Megan’s argument, additional error-making propensity has positive individual and social returns at the margin, which means we should offer implicit or sometimes explicit subsidies to experimentation.  That means a liberal bankruptcy law, relatively easy divorce, and a focus of subsidies on areas with high fixed costs.  In other words, the National Science Foundation should worry more about labs and less about individual researchers.

As I had predicted, this is proving to be a breakout book for Megan.  You can buy her book here.  There is more on the book here.


When SHAME does a write-up on you, you're generally doing something right, but from what I gleaned from her new book, I disagree with her message. Failure sucks your money and energy and as you get order it becomes harder to rebound. This is the reason why kids shouldn't take the libertarian approach of sipping college to pursue their own activities, because of the crippling effects of failure and the difficulties of rebounding. Getting a higher education gives you a foundation for success later in life; without it you are at a substantial disadvantage. Choosing to fail means others will get ahead. Generally, failing may work if it doesn't cost you much money or time or if it doesn't jeopardize a more important priority.

That's not the libertarian approach BTW.

I'm a market libertarian in that I believe in low taxes, free trade, and deregulation but I don't think it works so well at the personal level. A large corporation with next to nothing borrowing costs is better at handling failure than a person.

I think I see your meaning but that first sentence is a doozy - libertarianism is great just not at the individual level.

To work with the analogy, I'd look at whether subsidizing a corporations' borrowing costs is in societies interest over the full economic cycle.

My confusion - possibly mirroring Andrew's - is with the idea that "skipping college to pursue your own activities" is specifically or specially "libertarian".

I can't think of any principle of libertarian theory in general or economics in particular that suggests that as a blanket solution.

(Indeed, it might be a wise choice sometimes, for some people - and a terrible choice for others, at other times. But that all depends, not on "libertarian or not", but on the costs and likely benefits in the specific instance, does it not?)

Adolescence is the transition from childhood to adulthood, and the learning curve is steep. Wise parents let their children make decisions that they are capable of making, and to live with the consequences. Spend all your hard earned money on a trashy trinket, bad mistake, next time you do differently. To be clear, not allowing the kids to screw up their lives, but to allow them freedom that they can handle with their age and maturity level.

At one point in your life you have to start doing it.

"relatively easy divorce": oh dear, I hope that isn't bad news for the Official Blog Spouse. Or perhaps only the happily married can say such things comfortably.

The question is what she means by "relatively easy divorce". If she wants men to bounce back from divorce - and they are the main economic actors - then an easy divorce would be one that was not punitive. It is hard to bounce back from 18 years of child support. It is hard to bounce back from the loss of a house. If we want men to "experiment" with marriage, then the penalties for failure will have to be a hell of a lot lighter.

Perhaps she means that divorce ought to be more like bankruptcy? Admit your mistake. Get a few years to get your act together. Start with a clean slate. That would encourage more men to experiment with marriage.

The problem is not child support. The problem is the automatic award of child custody to mothers and the culture generally which provides a conduit for expressive divorce by both parties and actively encourages it among mothers. If McArdle's views have been adequately rendered by the moderator (and it appears more his gloss than her views), they do not take into account that people may be risk averse because marriages are evanescent and gratuitous penalties are imposed on fathers. We already have unilateral divorce on demand, so its simply bizarre to be advocating 'relatively easy divorce'.

But here's the thing: we're in a tale of two economies. There's all this crazy government data and people getting fired, and the Fed is printing, people are getting richer than ever, and AHHHh, it's scary. And I agree it is. But it's a great time to create wealth, to be wealthy, to be creative, to be smart, etc

My god this is asinine even for you...
Can you pleased tell us when it WASN'T a great time to be wealthy, to be smart, etc?

For the wealthy part, Russia/Soviet Union, the decade between 1917-1927 comes to mind.

Fair enough. Paris in 1789 as well.

Cambodia in the early 70's

Another one. OK, so your statement that we are better off than Cambodia in the 70s is quite accurate.


Would a socially sub-optimal aversion to failure imply that speculative bubbles may actually be a social good? Bubble-driven boom times allow the broader economy to coordinate to loosen how easy it is to succeed. All the crazy/marginal ideas and ventures that wouldn't get funded or even tried get the chance to see the light of day. A few may actually turn out to be quite good in retrospect. The bubble's burst clears out the rest that don't succeed and returns the economy to normalcy.

Interesting point...
Do note that among nations the US has culturally the least aversion to failure, meaning we are the most entrepreneurial culture, that's our killer app. Anything that helps keep that ethos alive should be top priority.

If by aversion to failure you mean the peace of mind bailouts at the corporate level and dischargeable bankruptcy debts at the personal level provide.

Definitely the latter, and kind of the former too. It shouldn't be the end of the world if your business idea fails, or no one would want to try. It's in our cultural DNA here to try stuff, and to not be ashamed if it doesn't work. Everywhere else, places like Japan for instance, failure is dishonorable, and the culture lacks entrepreneurial energy.

While I agree that "the National Science Foundation should worry more about labs and less about individual researchers",

I find it interesting that industry typically takes a libertarian stance on taxation, and yet, expects the government to pay for all research.

Industry, all industry, is non-ideological. I'm personally against state run gambling parlors. If the state comes to me tomorrow and offers to let me have a dozen slot machines as long as we split the take, I'm all over it. I may still vote (if I still voted) for people promising to get the state out of the gambling rackets, but I'll take the deal while it is offered. The Koch brothers are not turning away any subsidies on ideological grounds.

Maybe at the fringes you find companies serious about ideological issues like ecology or other religious issues, but in the main business is about business, not ideology. It is why liberal politicians run circles around conservative politicians. The former grasps this reality while the latter thinks business is his ideological soul mate.

Like I said, at the fringes you get this stuff, but in the main, business act like businesses.

Well perhaps you live on your own fringe then. For the rest of us, Smith's "desire to be praise-worthy" seems a universal. See also "the Gates Foundation."

Bill Gates = one guy

Microsoft = lots of guys

The difference? One tries to buy a place in heaven the other sells software.

You really should not try so hard to be wrong.

Do I need to try hard? What did you even just claim, that you know the total charitable contributions out of MS, and it is one?

Also, Smith's "desire to be praise-worthy" is definitely not the same as a place in heaven.

Man naturally desires, not only to be loved, but to be lovely; or to be that thing which is the natural and proper object of love. He naturally dreads, not only to be hated, but to be hateful; or to be that thing which is the natural and proper object of hatred. He desires, not only praise, but praiseworthiness; or to be that thing which, though it should be praised by nobody, is, however, the natural and proper object of praise. He dreads, not only blame, but blame-worthiness; or to be that thing which, though it should be blamed by nobody, is, however, the natural and proper object of blame.

For what it's worth, I think most scientists see a "big science" vs "small science" divide, rather than any "labs vs researchers."

Basically there are many good small projects, but like anything else that must pass Congress they must be carried by some big idea, big mission. Some small satellite project rides the tail of "mars mission funding," or some small research on disease rides the tail of "human genome."

On this, I was surprised to find (and be late to)

Drop a few bucks on small science.

This is an instance of a special interest group (Scientists, I'm looking at you!) having interests that are at odds with those of the general public.

There's some argument for the government funding big science - things that can't realistically be funded by industry, foundations, and other private sources. It is much less obvious why the government should be funding small science which, were it important to people and industry, could easily be funded by them.

I think your perception is correct, and scientists see it the way you described. But they are wrong if they think the correct solution is more generous and less accountable funding of small science. If a project is languishing because it can't raise a couple of million dollars, that's a powerful signal that maybe it's not worth doing. By any historical standard we are a fabulously wealthy society awash in money. If you can't find a few million for a good idea, it's not actually a good idea or your ability to execute it is in doubt. If a project is languishing because it can't raise $16 billion, well, that doesn't give you nearly as much information on whether or not it's worth doing. There might be a case for the government funding it.

In college, I worked on a project looking at mass interventions for accidental prescription drug usage by children (think safety caps, except for preteens). It cost half a million dollars; the NIH quite happily paid for it. There aren't a hell of a lot of private foundations for this stuff (and those that exist are way more competitive than the NIH grants, or they were at the time).

The societal benefit-cost ratio for this project was incredibly high, but what does Pfizer care about that?

I think you misinterpreted me, and are wrong to boot! I think the majority of scientists have small-science projects. Their "goodness" is not necessary better than big projects on, average, but you can do a thousand of the small projects for the price of a big one. With small science and small failures you are playing the odds, without the risk of single ticket item billion dollar boondoggles (I'm looking at you, Hydrogen Highway).

No, I think the boosterism for big science comes from the increasingly political layers above the lab, starting with lab directors and ending with lobbyists to legislatures.

There is a difference between recognizing that a certain amount of failure is a natural consequence of an optimally experimental approach to certain activities, and celebrating failure outright. So, if you are an inventor or entrepreneur you should be open to a level of creative experimentation that will naturally have a high failure rate. I'm not sure that this applies to all activities though - let's say being a practicing doctor or structural engineer - where there is a large body of useful knowledge and the risks of being wrong are large. Society and all it's activities and institutions are not one large collection of constantly experimenting entrepreneurs. Creating "labs" (broadly defined) for low risk experimentation is good, but there is also a time when the "product" needs to go into production. Dating is fine; getting married five times maybe less so.

How do any of these endeavors where risks of being wrong are large know what is wrong? It is because someone did it, and it failed. There is no way to prevent it, people are not machines. Yes there is a body of knowledge, and it is taught, but at one point someone has to get practice and experience. Experience has been defined as someone who has made all the mistakes, and believe me, that is the case in structural engineering or doctors. Smart managers of these people let them gain experience in situations where the stakes are not high.

The point is that it is impossible for failure not to happen, and the more onerous the structures to avoid failure, the less that will get done, the less growth that happens, the less new ideas that come to fruition.

And the more costly it is to admit a mistake, the higher the costs of doing anything.

And yes, sometimes the stakes are high. Make it that stakes are not that high. If, for example, the economy of the US depends on the wisdom and experience of one person who is the Fed Chairman, or the CEO of one of a few large banks, then the stakes are too high. Mistakes will occur, so make it so that the whole thing doesn't collapse when they inevitably happen.

I think Megan M is the XX version of Malcom Gladwell. From the preview of her book, apparently a bunch of kindergarten kids beat a bunch of engineers in a contest to build the highest structure using pieces of uncooked spaghetti and tape that would support a marshmallow by simply asking for more spaghetti and placing it in an unstructured pile, while the engineers build an elegant suspension structure that was a bit smaller (apparently the MBA students and lawyers did the worse, as they were arguing who would be in charge). This is bogus logic as it rewards rule breakers. OK I would beat even the kids: I would tape the pieces of spaghetti in a string, and suspend the marshmallow from the end of the spaghetti string, then argue that you must measure distance not from the ground up but from the ceiling down. That makes me some sort of genius? Well, come to think of it, given that I am a genius, maybe Megan M has it right after all. ;-)

Gladwell has a proven market model, a design pattern. I don't really fault McArdle for following that, nor even "defending failure," though that defense seems kind of stale in 2014. We all know try, try, again stories ... going back to Thomas Edison. So, I'd prefer a book with a preface on failure, and chapters on strategies in this age. Though, that isn't really the Gladwell pattern.

Few, if any, engage in an unbiased search for truth. That's why Western legal systems start with the assumption that all involved are liars. MM is trying to sell books. She takes a well accepted belief, "you learn more from failure than success," and wraps a bunch of pseudo-science around it. People like being told they are smart almost as much as they like being told they are right.

Remember the studies ... students do better on moral questions before taking economics. An over-emphasis in the latter degrades other aspects of human nature. See also "handling money."

I watched the video last week and enjoyed it very much. McArdle's thinking has definitely evolved over the past few years but she is, at heart, a libertarian. I like to say that the best motivator is fear, and I say it from personal experience. To Carney's credit, he did ask how the upside of down applies to bankers. Cowen responded with a few handwaves, and McArdle responded with a discussion of Dick Fuld. I suppose for a few the upside is higher than the down, and for many the down is much lower than the upside. Like I said, fear is the best motivator.

relatively easy divorce

"The further you go, the less easy it will be to withdraw; yet no oath or bond is laid on you to go further than you will. For you do not yet know the strength of your hearts, and you cannot foresee what each may meet upon the road."
"Faithless is he that says farewell when the road darkens," said Gimli.
"Maybe," said Elrond, "but let him not vow to walk in the dark, who has not seen the nightfall."
"Yet sworn word may strengthen quaking heart," said Gimli.
"Or break it."

Lax divorce law. Just what we need.

I think her point is mainly with alimony, where some states are really behind the times. My state of MA had a woefully antiquated view of the alimony requirement. You could be married quite young for a couple years and be saddled with a lifetime of alimony payments. Not only might this discourage any same man from getting married, it made remarriage a dicey proposition (your spouse's income was now included with yours).

Change finally came - see this
These are sensible reforms (despite the slant of the post).

Mass has one of the more insane child welfare operations on the planet. Years ago I was in a business there and we had a fair number of entry level employees. These were blacks and Hispanics mostly. The turnover was steady because the state would call up about six weeks into their employment demanding we hold the paycheck for child support reasons. Of course, the guy would quit. Any attempt to reason with these state bureaucrats was met with threats. The whole division appeared to be staffed with bitter, divorced middle-aged women.

Amen. Even my generally pro-feminist spouse thinks these policies are a disaster.

And despite their outlook, the level of incompetence is sometimes astonishing:

One problem that can arise is that child support orders are derived from 'imputed income' and can be difficult to modify. Fathers are often in a wretched position if their circumstances change.

Some of your employees might have been advised to keep their pants zipped. There are a great many divorced fathers who have been dealt with very shabbily by the courts, but most people paying support these days are likely those who sired bastards because they were libidinous and careless.

The whole division appeared to be staffed with bitter, divorced middle-aged women.

I suspect people who are attracted to social work as an occupation and slog through the vapid course material are disproportionately drawn from the world's problem people. I've mostly dealt with hospital and nursing home social workers re elder care, and they are uniformly fifth wheels. The one amongst my proximate relations eventually drank herself to death.

One plaintiff in 25 is granted alimony. There may be some individual injustices in there, but it is not a pervasive social problem.

Partway through her book right now (sorry, Megan -- I checked it out from the library, where I got off the hold list on Sunday, so you won't get royalties from me). One wonders how her book would have been different if she had children, since there's no bright line between "failure they can learn from" and "failure that they need to be protected from, because they're just not able to figure things out in an adult way yet." And the difficulty that I'll be interested in seeing how she handles is this: even if it's good for society for people to take risks, fail, and learn from their failures, nonetheless, there are people who get back up, and people who simply are never able to. Is failure (i.e., risk-taking) necessarily good for individuals?

She married quite late in life (at 37) and is likely too old now for children. If I am not mistaken, she is an only child, so sad for her mother and father.

The Dalai clique has been treated coldly frequently while it is still willing to be a pawn for the U.S, it said. Over the past 60 years, the Dalai clique has colluded, taken advantage of and supported each other and internationalized the "Tibetan issue," the article said.

Comments for this post are closed