Broken Symmetry

One reason why academia does not appeal to me

If you’ve got someone interested in going across departments, spending part of the time in physics and part of the time elsewhere, their colleagues are going to kick them out. They’re not contributing fully to any single department. OK, every so often you’re going to get a huge benefit, but from day to day, where the universities are making appointments, they want the focus in one field.”

From Edward Carr via Marginal Revolution

Do Good Fences Make Good Neighbors?

Why, yes, they do.

There's a debate among academics in patent law now over whether patents create an anticommons (or "gridlock") problem. The debate tends to assume without proving that there would not be a worse commons problem were patent rights peeled back to disable the system of exclusive rights to technology that developed in the 19th century in the United States.

But patent rights weren't the only property rights that were secured for the first time in the 19th century United States. In fact, the invention of barbed wire (itself patented) contributed to a dramatic decrease in the costs of securing real property in the 19th century. And to what effect? Richard Hornbeck reports:

This paper uses the introduction of barbed wire fencing to the American Plains in the late 19th century to estimate the effects of property rights on farmers' production decisions. Farmers were both formally and informally required to build fences to secure their land. From 1880 to 1900, the introduction and universal adoption of barbed wire reduced the cost of fencing, relative to wooden fences, most in counties with the least woodland. Over that period, counties with the least woodland experienced substantial relative increases in the improvement intensity of farmland, the value of farmland, and the productivity and production share of crops most in need of protection. This substantial increase in agricultural development appears to reject increased security of property rights due to barbed wire fencing. Some states' previous efforts to reduce the need for fencing are not found to have had similar effects, however, suggesting a difficulty protecting property rights in sparsely settled areas through legislation and the court system.

Why would state-enforced rules not work as well? See Robert C. Ellickson and Coase. (Still not transparent? See Calabresi and Melamed.) And note that patents are more like barbed wire than like state legislative efforts to shift the costs of fencing to herd owners (described in the paper).

If this seems like ancient history to you, then consider the fact that our government is going to have to learn how to be 100x more efficient with its tax revenues over the next few decades if we are to avoid insolvency.

Here's another interesting point: Ellickson found that even barbed wire was too expensive to be worthwhile in close-knit communities of ranchers and farmers like Shasta County, California.

Economics that can be Falsified

From K. Bastiaensen, P. Cauwels, D. Sornette, R. Woodard, and W.-X. Zhou.

Amid the current financial crisis, there has been one equity index beating all others: the Shanghai Composite. Our analysis of this main Chinese equity index shows clear signatures of a bubble build up and we go on to predict its most likely crash date: July 17-27, 2009 (20%/80% quantile confidence interval).

Available here. Found via Physics arXiv blog. Can a reader tell me who authors and edits this blog?

Didier Sornette's work on the endogenous and exogenous trends in YouTube video watching was covered in this earlier post.

The argument here rests on the observation that faster-than-exponential growth is not physically sustainable. A similar point was made in an earlier post on this blog regarding venture capital financing and startup growth: The Fundamental Speed Limit on the Growth of a Startup.

UPDATE (August 2, 2009): Make that, economics that has been falsified. The Shanghai Composite blew through the intervaal identified showing few signs of slowing. Conway and Kochen trump Sornette et al.

Why is our System of Executive Compensation so Broken?

Robert H. Frank has an answer:

When the strength of positional concerns differs across domains, the resulting conflict between individual and social welfare is structurally identical to the one inherent in a military arms race. To illustrate, consider rival nations faced with deciding how to apportion available resources between domestic consumption and military armaments. Each country’s valuations are typically more context-dependent in the armaments domain than in the domain of domestic consumption. After all, having lower domestic consumption than one’s rival might entail psychological discomfort, but being less well armed could spell the end of political independence. The familiar result is a mutual escalation of expenditure on armaments that does not enhance security for either nation. Because the extra spending comes at the expense of domestic consumption, its overall effect is to reduce welfare. Note that if each country’s valuations were equally context-sensitive in the two domains, there would be no arms race, for in that case the attraction of having more arms than one’s rival would be exactly offset by the penalties of having lower relative consumption.

So long as CEO pay performance is determined by what other CEOs are getting paid rather than by how much value the company creates during the CEO's tenure, CEO pay will be a positional good in Frank's terms, which draws value away from employees, customers, and investors.

How could CEO pay become less positional? By giving other stakeholders (but especially investors) more transparency into the cash-flows in and out of the firm. Boards and shareholders are forced to compete with other boards and shareholders because those measurements of CEO performance are right now more reliable than more direct measurements of company growth.

Sen on Smith

Trenchant commentary on the crisis:

Smith rejects interventions that exclude the market—but not interventions that include the market while aiming to do those important things that the market may leave undone.

For earlier commentary on a similar theme see here and here.

Do we really want the American auto industry to declare bankruptcy?

There's been so much commentary on the auto industry over the past month, but almost all of it focuses on how letting the industry "fail" by falling into bankruptcy is somehow necessary to the healthy functioning of the economy. The consensus seems to be that while a bailout of the financial industry was necessary because everything else in the economy depended on banks for access to credit, a bailout of any industry that actually makes things (like the Internet industry in 2001 or the auto industry now) is both unnecessary and undesirable because failure would facilitate a quicker reorganization of labor, management, and creditors into a manufacturing company that actually makes things that consumers want.

I have no trouble with the theory behind these arguments, which is basically Schumpeterian or at least Austrian. But I want to point out a few things that I think have been under-appreciated in other commentary:

• Bankruptcy is a lose-lose situation for all stakeholders except (in principle) consumers. Senior creditors often want a bankruptcy when they believe that they're only going to get fewer cents on the dollar by letting management continue operations, but over the long haul, even senior creditors do better if the company can pull through a short-term liquidity crisis. I say consumers in principle because it depends on whether the new management that often takes over in bankruptcy is able to do any better with the economic problems that a company has to solve than the old. There are reasons to be skeptical that anybody could have done much better over the last decade.
• Wiping out shareholders basically makes raising capital for new entry in the industry much more difficult. Although debt might be easier to raise from senior creditors that were able to hedge their downside through forced bankruptcy, it won't be easy until stable revenues are coming, and that's not going to happen without significant capital expenditures and time reorganizing management and labor.
• Assume GM and Chrysler are forced into bankruptcy in March 2009. If Ford cannot get the same deal that GM and Chrysler get with the UAW at that time, then Ford will effectively be suffering for having been better managed.
• Bush is looking to protect his legacy, and simultaneously would have been giving Obama a pass on a thorny political issue by executing an auto industry bailout during his lame duck term. Too few commentators are considering how the announcement on Saturday represents, in effect, a punt to the Obama administration on the toughest issues.

Patent Thickets Revisited

The VOX EU blog carries this excellent summary of empirical research done by Mark Schankerman and Alberto Galasso. In studying patent enforcement, they arrive at several conclusions:

• Establishment of the Federal Circuit facilitated dispute resolution by clarifying property rights.
• After controlling for other relevant factors (including measures of patent value and age, the involvement of serial infringers and patentees, and other factors), patent suits do indeed take less time to settle when the alleged infringer is using patents in technology areas where ownership is more fragmented.

That's right. According to their empirical work, patent thickets actually decrease the duration of disputes over technology because less is at stake in each negotiation in a fragmented market.

It would be interesting to hear Gridlock Economy author Michael Heller respond to this work. According to Schankerman and Galasso at least, overall transactions costs are not increased significantly by fragmenting patent ownership.

Are short time horizons reinforced by the media?

I'm not a CFA or registered broker agent, so this isn't investment advice. But have you noticed that at the same time that short-term treasuries have been at historic lows, the long-term treasuries are very low also? Sure, the story about 1, 3 and 6 month t-bills makes for great headlines -- those are dramatic cliffs in mid-September. At the same time, there hasn't been the volatility in the longer-term notes. 30 year notes, for example, are still hanging around 4 percent, where they've been all year.

Which demonstrates quite vividly that nobody really believes that the equity market isn't going to resume a normal, sustainable (8% to 15% depending on how optimistic you are politically) rate of growth over the next 30 years. Yet most investors are not ready to jump into the equity market for the long-term (as the short-term rates ably demonstrate).

Again, I'm not an investment advisor or macroeconomist by training, but the money being used to buy short-term notes will have to go somewhere within the next six-months when those short-term notes expire. Real estate, like all assets whose value isn't fundamentally derived from human capital, doesn't have the growth potential of equity. Debt markets will surely pick up for some high-quality issuers, including the manufacturers of staple commodities. But for long time horizons, equity is looking better and better as 2008 goes on. Besides that, selling equity now means you fail a basic test of investing -- if Warren Buffett is buying, are you buying or selling?

This is not healthy for lawyers or clients

What you're looking at is the bimodal distribution of starting lawyer salaries.  If you integrated up the area under the right lobe, you'd figure out pretty quickly how many lawyers are working in "big law," and if you compared that number to the number graduating from top-U.S. News-ranked law schools, you'd get a pretty good idea of where in the U.S. News rankings your chances of making the big money fall off dramatically.

This distribution can be interpreted in a number of ways.  But perhaps the simplest explanation is that the demand for legal services from large corporations is concentrated on the supply for a handful of top law firms.  Why is this?  Because it's very difficult to measure the quality of legal services.

How could you change this?  For starters, you could change the rules of professional ethics that prevent lawyers from partnering with clients to provide services.  If lawyers and clients could share equity, their interests would be more aligned, and billable hours and pedigree would be less necessary as proxies for quality.

Via Greg Mankiw

How to Improve Traffic Flow

I was driving on Highway 17 on my way to Santa Cruz today, and I started reflecting on the question of how we could improve traffic flow using principles of synchronized flow.

Often on the highway, one will observe a slow driver in the fast lane, holding up a line of traffic.  This is particularly problematic on two-lane roads.  In physics terms, lane changes increase the viscosity of flow, which in turn slows the average rate of traffic, and increases the probability of accidents.

But because humans are driving, we have the possibility of coordinating behavior through incentives.  Ten years ago, this would have been impossible, because two drivers had no way to communicate efficiently (at least not safely) on the freeway.  Now we do.  Here's a very rough idea of how to solve the problem of highway traffic with incentives and technology:

1. Sell an RF device to drivers that can be installed within easy reach of the steering wheel, with an easily visible and readable display that registers a unique identification whenever another such device is nearby -- say within 100 meters -- for example, by displaying a picture of the make, model, and color of the car.  Make it easy to cycle through the nearby cars on the display.
2. Give each driver who installs the device the option of loading a certain amount of cash into memory, say $100, with a specified "tip unit," which could be decided by the user (say, $1).
3. Whenever a driver with the device installed comes up behind another car with the device installed, the device can prompt the driver behind: "Offer Tip?"
4. The driver may then press a button, which signals the device in the car in front that a tip of $1 is being offered from the car behind.
5. Now if the driver in front pulls over, the driver behind can press a button that says, "Give Tip."
6. The device in the car behind then executes an electronic transfer to the device in the car in front.   People who are not in a hurry can literally earn money for accommodating drivers who are.
7. By keeping each device synced with a mainframe by cell connection, the company selling the device can maintain a rating for each tipper (so that you can't game the system by promising a tip, but then not executing) and other anti-fraud measures.

Now obviously this is somewhat regressive as a "tax" on driving time because some people are going to be able to make much bigger tips than others, &c.  But I think it's still a net benefit to society because it would decrease accidents and road rage, and probably increase the average flow of traffic by a lot.  Right now, it's impossible for anybody to know whether the guy in the car behind them is just a jerk in a hurry or whether they're on their way to the hospital to save another person's life.  Besides, it's also a progressive "tax" in the sense that you don't have to pay anything if you take public transportation instead.

If anybody knows of any startups working on a similar idea, I'd like to know.  Because of the incentives of picking up free money and shorter commutes, I bet it wouldn't be hard to sell these devices.

What the Design of SUVs and Cost Accounting have in common

The design of both was predicated on an unlimited supply of cheap fuel.

When SUVs were designed, the costs of fuel were low relative to the joy that people got from driving them relative to other kinds of cars.  Nobody cared about miles per gallon.  What they wanted to know was how big and how fast it could go.  Don't get me wrong.  I love cars that growl and shake when you start them.  But that growl and shake come literally at the expense of fuel efficiency.  The design loses its appeal when the price of fuel starts to cut into my enjoyment of other things.

Similarly, cost accounting was designed to measure and report activities within a firm at a time when "fuel" (i.e., capital, labor, and raw materials) were very cheap relative to the demand for its products.  Globalization supercharged that trend, and kept the flawed model of cost accounting on life support.  When labor got expensive, we simply moved to another part of the world where it was still cheap.  Capital got cheaper!  Only raw materials were getting more expensive.

And here we are in 2008, with the number of places in the world where labor is significantly cheaper than it is in the United States shrinking by the day, with capital getting more expensive thanks to inflationary monetary policy, and with raw materials more expensive than ever thanks to an oil shortage.

Time to rethink our theory of accounting folks.  It's not how big or how fast you are anymore.  It's your miles per gallon that matter.

Why don't VC funds employ in-house lawyers to do work for portfolio companies?

A while back, Jason Mendelson wrote a great post about his frustration with "startup lawyers," by which he means outside counsel who work on transactions relating to portfolio companies.  Like many consumers of legal services, Jason is frustrated with a trend toward lower quality and higher cost.

I don't want to get into the discussion of what trends have emerged and why here.  Rather, I want to ask the question of why more VC funds haven't hired outside counsel to come work in-house doing work for portfolio companies.

Benefits:

• Lower fees from outside counsel, probably more than offsetting the salary cost
• Better quality services from outside counsel (higher bandwidth communication, especially when it's with a former colleague)
• Less worry about the billable hour clock by the in-house lawyer, which means more efficient allocation of time to various legal problems
• Lower cost to lawyer spending time on-site with portfolio companies to proactively avoid legal problems
• Lawyer also available for consulting on fund-related issues when not needed at portfolio companies

Disadvantages

• Requires outlay of cash-flow from management fees to cover salary
• Agency costs that could result from soft kickbacks between in-house lawyer and outside counsel (relatively easy to reduce with monitoring)
• Misalignment of incentives between lawyer and portfolio companies because salary is paid by VC fund (but this one is shared with the VCs!)

Note that nobody could eliminate outside counsel because of the superior access to information about new case law, unusual market scenarios, the database of internal legal documents built up over years for use in a variety of client matters, lower cost ability to generate stock ledgers, etc.

Often, VCs invest millions of dollars into a new round of financing, only to demand that the startup company turnaround and pay outside counsel for the tens of thousands of dollars in costs of advising on the transaction.  I understand why this is done from an accounting perspective.  But it is not so cost efficient.

Both an Agency Cost and an Externality

From VentureBeat, a quote from a Cisco VP on how and why virtualization doesn't always result in power savings for companies running large data centers:

"The reason a lot of virtualization practices never show up as savings is because the facilities department, which is not part of IT, will do what they think is right: Install processes on their cooling system that add an artificial load in direct proportion to the load you saved in the virtualization process. The device they install is called a hot-gas bypass. They take the output of the cooling system and pipe it into the input. So it keeps the chillers operational."

You couldn't make this stuff up.  It goes to show that you need good engineering in both your institutional design and your capital equipment if you're going to make cost improvements.  Focussing only on one or the other won't always help.

Patents as Intangible Asset Partitions

Before amending the patent law, Congress should consider the incredibly expensive alternative to our patent system: a world in which only trade secrets and contract rights can be used to define and enforce rights to intangible assets, such as R&D.

Some relatively recent academic work on the law of corporations has discovered a point that was probably obvious to many business people all along: that the corporate form is often more useful as a shield of the corporation's assets from shareholder creditors than it is as a shield of the shareholders from the corporation's creditors.  The terminology adopted to describe these different uses of corporate limited liability are "affirmative asset partitioning" and "defensive asset partitioning," respectively.

A key insight is that shielding a corporation's assets from each shareholder's creditors (including descendants and donees) permits a more stable pool of capital to be developed and deployed.  Without the corporate form, shareholders would have to place company assets beyond creditor reach by contracting with each creditor -- a project too expensive to be worthwhile, especially when the benefits of the contracts are shared with the other shareholders.  The corporation as an asset partition dramatically reduces the transaction costs of forming the pool of capital necessary for putting it into business.

What do asset partitions have to do with patent law?  In a very clever paper on the subject, Prof. Paul Heald argues that patents serve a similar transaction costs reducing function in pooling the human capital necessary to complete projects within a company.  To see how, consider that -- without patent rights -- companies would have to rely solely on contracts and trade secrets to protect their R&D.  Unfortunately, because employees can't willingly forget what they learned doing R&D at Company A, there is no practical way to enforce contract and trade secret rights without taking extraordinary (and expensive) measure to prevent employees from leaving, taking with them to Company B the family jewels.

Business history is replete with stories of raids of R&D employees.  For companies, and even whole economies, the essential ingredient of economic growth is human capital.  An underappreciated point about patent law is that it permits the more efficient development of human capital by a more free-market for R&D employees and less risky joint-venture among companies.

A corrollary is that if patent rights are weakened too much, companies will do redundant, wasteful R&D in producing the same level of innovation for consumers -- and if it's really valuable innovation (i.e., with highly inelastic demand), it's the consumers that bear the costs.

Congress and Courts should bear this in mind when trimming patent law back.

Hockey Helmets and Patent Markets

Years ago most hockey players did not wear helmets. Yet when privately polled, most hockey players said they would prefer to wear helmets for the extra safety that helmets would provide. Why the difference?

Economist Thomas Schelling pointed out the problem: when almost no player wears helmets, the player who does puts himself at a disadvantage in terms of peripheral vision. Although each player preferred the saftey of helmets, none was willing to accept the disadvantage in exchange for the extra safety. The answer: make helmets a league requirement. When everybody wears helmets, everyone is safer and everyone at the same disadvantage.

Patent buyers and sellers are in a similar position to hockey players right now. Although plenty of comparable patents are being licensed and assigned at any given moment, and public information about the terms of these licenses and assignments would make valuations considerably easier for everyone (saving some legal and expert witness fees at least), nobody alone would dream of doing so because of the advantage this would give competitors in terms of estimating costs of development and competition.

The soluton for patent buyers and sellers is as simple as the solution for hockey players -- require ALL patent license and assignment agreements to be disclosed. Everyone benefits from more accurate valuations. Nobody is at a competitive advantage in terms of disclosure.

Lemley and Myhrvold proposed this in an oped earlier this year.

Collective Action Problems on the Web

Marc Andreesen, whose own foray into blogging has more or less inspired me to follow suit, remembers Dave Sifry explaining how blogs would help solve the collective action problems of spam and comment trolls.  In response, Tyler Cowen predicts the future of blogging to look pretty much like the status quo.

Blogs do help solve collective action problems by centralizing decisionmaking into the hands of one or one group of bloggers, who have an incentive to keep their webpage as hospital an environment for readers as possible.  The Chicago school answer to collective action problems is to centralize decisionmaking by assinging exclusive rights (i.e, "property rights") to the highest value user.  Because most blog software permits non-anonymous bloggers to internalize the positive social benefits of mindshare from blogging, bloggers will tend to remove value-destroying comments and advertising on blogs.  The negative costs of spam are spread over the entire Internet.

But there's nothing particularly special about blogs in this regard.  Any software platform that bundles exclusive rights to content and assigns it to a particular person or group would achieve this same benefit.  A while back I saw that the number of new blogs had leveled off as social networks such as MySpace and Facebook took off.  This is most likely because for some people social networks are an economic substitute to blogging.  (For others, such as Tyler Cowen, who recently setup a Facebook profile they may be economic complements).

As a competitive matter, I believe this cuts in favor of Facebook, which has maintained far more control over the content presented on its pages than has its competitors.  Indeed, Facebook privacy settings could be viewed as a means for fine-tunining the risk-reward tradeoffs between the benefits of more widespread dissemination of personal information on the Internet and the costs.