Failure must always be an option | Hacker News Reader

Failure must always be an option

rondam.blogspot.com

63 pointslisper23 comments

While I believe this article is a fascinating look at the bureaucracy at JPL, it is analogous to the mortgage crisis only in that it is an example of perverse incentives. Perverse incentives exist all around us, so I don't find this example particularly enlightening.

Some other examples of perverse incentives in systems: http://en.wikipedia.org/wiki/Perverse_incentive

Right, in the economic case, the perverse incentives were created by the market. The bailouts are only meant to stop the failure of our economy as a whole. That sort of failure must never be an option.

not created by the market. banks took these high risk loans knowing they could sell them to fannie-mae and freddie-mac, who in turn knew they would not be allowed to fail.

Actually, weren't lending institutions required to loan so much to high-risk borrowers as a way to make sure more people had homes? Seems like I heard this was a requirement if the institution wanted to open new offices.

The loans themselves are only a tiny fraction of the problem. It's more the credit default swaps and other forms of gambling tied to the loans.

http://en.wikipedia.org/wiki/Parable_of_the_broken_window

What does all this have to do with the mortgage crisis? I believe that the mortgage crisis is essentially my experience at JPL writ large. Both situations were produced by a set of rules that produced perverse incentives that rewarded people for acting contrary to the greater good. In my case, I was rewarded for publishing useless papers. In the case of the mortgage industry, banks and brokers were rewarded for making bad loans. And in both cases, if push came to shove, the government with its deep pockets was there to foot the bill.

when you hear the truth it rings like a bell.

Is there really a moral hazard here?

Does the bailout really remove disincentive for failure? Sure it lessens it, a bit. But this article is describing a gross moral hazard. I don't think this is comparable.

AIG shares have lost virtually all of their value. The bailout rescues some of it, but it's probably less the 5%-10% (please correct me if I'm missing something). Imagine AIG's morally hazardous executives a year or more ago (share price = $65 - $75). They know that while their desicions could cause a collapse, there's a $3.50 floor. Does that have much of an effect on their desicions?

What this article shows is how "moral hazards" & close relatives exist within organisations to a much larger extent then they do between them.

There are definitely a lot of perverse incentives that lead up to this, but they seem to be largely the fault of private enterprise.

If an executive's compensation is largely based on options, then every share price at the time they come due that is below the strike price is the same to him. So if he's given the option to purchase shares for $50 each in a year, then he makes the same if his company is trading at $49 as he would if it were trading at $2.

Add in a golden parachute and he's in a situation where his incentive is to gamble. If things go well (for instance, if people with bad credit somehow paid their mortgages consistently) share prices might be at $100. And if they tank, well, he gets his $12 million and floor seats at the Knicks games for the next 5 years, which is the same thing he'd get if things only went a little down hill.

There are definitely all sorts of problematic incentive structures all over the place. And they're definitely a problem. I think that just goes to show they're inevitable to a degree, that they're survivable & that free markets are not a silver bullet. These bailouts are certainly a push in the wrong direction, but unless I'm missing something, they're a very minor push. As you say, options as incentive are worse.

Compared to the other effects, that doesn't even show up on any kind of radar. For one thing, these companies are now essentially nationalised. Just the implications in terms of international politics & rhetoric are pretty big.

What did nationalisation do to the incentives of shareholders of other too-big-to-fail companies? What did it do the the execs. From the 1 year ago perspective this is still failure. 'This stock might lose 100% of it's worth' & 'this stock might lose 93%' look very similar when you are sitting on 100%. If any kind of serious distortion happens it probably happens near the bottom when things are pretty chaotic anyway. Executive compensation might be causing distortions, but the US' new found nationalisation policy has nothing to do with that.

Netcan, if this exercise does anything it should prove to most that free markets do work and one of the reasons they work is that they always have lots of "silver bullets" to spare. What we are watching now is silver bullets riddling companies and company execs who probably need to be riddled with silver bullets.

Unfortunately some innocents will end up going down with the ship but this has always been the case. Events like we are watching right now are caused by human weakness and human failings not by free markets.

I'm not an expert on the mortgage crisis. I've had a lot of experience over the years "fudging the numbers" in the real estate sector but that was before I became a web developer and, thereby, an upstanding example of responsible personal economic behavior.

I am an expert on Matrix Management though. Unfortunately, I have lots, and lots, and lots of experience there.

My recommendation..... If any of the big shots at your company say that they are switching to "Matrix Management" just polish up the old resume and start looking for a job elsewhere.

"months to years -- and in the meantime my line management still had to keep me employed somehow. On a number of occasions I had proposals rejected on the grounds that what I was proposing was impossible, when in fact I had already done the work in the intervening time."

That bit made me chuckle...

"And in both cases, if push came to shove, the government with its deep pockets was there to foot the bill."

Americans foot the bill. Government is just a group of people acting on behalf of Americans.

It's easy enough to get upset at the bailout, but the fact is that we Americans are going to pay the price one way or another, and this way is (or so the theory goes) significantly cheaper and less painful than a full-on depression.

The perverse incentives that caused our current economic crisis were almost entirely the fault of private industry left unregulated. So the same people who are angry about the bailout now would have been angry then had the government started intervening in the private markets by regulating private enterprise more tightly than appeared necessary.

The most important thing about the bailouts is that failure was not removed. Shareholders in many bailed-out corporations are losing tremendous amounts of money. It is sad to see the executives who ran these corporations fall gently due to their golden parachutes, but those were not the making of our government.

Oh really?

ML sold its mortgage portfolio for $0.22 on the dollar.

Wouldn't you have loved to have bought even for $0.25-30?

We're not going to see 50% forclosures and even if we did, housing prices are not dropping 70%.

The US govt shouldn't buy ANY of these securitized packages. Instead, they should be offered at public auction.

The collapse happened because govt limited the size of the market. When those folks got spooked, things fell apart.

If I'm not allowed to buy something for my benefit, I shouldn't be expected to pay for it for someone else's benefit.

I think you've got your math confused. One foreclosure can wipe out the profit from multiple non-foreclosed houses. It doesn't take 50% foreclosures to wipe that out, even if housing prices don't drop. With housing prices dropping the way they are, it might only take 10%.

This is a great post because it helps evaluate the very different goals of various organizations. That is, it shows that the "products" of large institutions (government, higher-ed, corporate firms) are not necessarily those of smaller businesses (start-ups, boutiques, small non-profits). It also helps us consider what exactly "failing" is (it has very different meanings if you're a hacking in your free time or applying for a CS PhD, even if the output is very similar).

So, in short, one man's "useless paper," may be another's inspiration. I'm not saying all products are the same, but it's important to evaluate context and think carefully about the complexity of a product's "worth." I mean, finding value (or being able to extract value) from "junk" is a goal for most entrepreneurs.

"At the peak of my JPL career I attained the rank of Principal, which is the highest rung on the technical career ladder whose existence is publicly known. (It turns out there are "secret" promotions you can get after that.)"

So, that's when you get to work on all the X-Files stuff?

Actually I think, that's when they let you in on the real secret, there is no "X-Files stuff" and leave you to wonder what the heck you just wasted the last 20 years of your life doing.

http://citeseer.ist.psu.edu/allcited.html

Who is "D. Johnson"?

David S. Johnson: http://www.research.att.com/~dsj/

Co-author, with Michael R. Garey (http://cm.bell-labs.com/cm/ms/former/mrg/), of "Computers and Intractability". (See http://citeseer.ist.psu.edu/articles.html .)

True scientists celebrate failures.

Great reading. I recommend it heartily.

Reply on news.ycombinator.com