How much do CEOs really make?

The Huffington Post observes that pay for "[t]he head of a typical large public company" was $9.7 million in 2012, a record high. But what really is the "typical" large company, anyways?

From the BLS, the mean Corporate Executive pay is $176,840, and the median $168,140. Obviously the well-paid outliers drive up the mean, which explains why it's higher than the median. However, the enormously well-paid executives are indeed outliers.

There are roughly a quarter of a million Corporate Executives employed in the U.S., and the vast majority are not seeing seven-figure salaries. The public furor over excessive corporate pay is an anomaly for the profession, not the norm. But depending what sample you look at, there is always something to feel outraged about.

Tornadoes and nuclear weapons

The recent tornado in Oklahoma is a great disaster, but the comparisons it has spawned are perhaps just as disastrous.

Fox News:

Several meteorologists consulted by the Associated Press estimated the tornado's energy released during the storm ranged from 8 times to more than 600 times the power of the Hiroshima bomb, with more experts at the high end.

True perhaps, but also highly misleading.

1. The tornado lasted 50 minutes, vs. a nuclear bomb going off instantly. In terms of "power per second", the tornado barely compares.

2. As modern nuclear weapons go, the bomb dropped on Hiroshima was tiny. It was 15KT, while the largest weapon tested was 15MT - a thousand times more powerful.

So yes, the tornado was bad, really bad. But comparing a natural disaster to a man-made catastrophe is rather deceptive.

Memorious Social Science?

I read the short story "Funes, the Memorious" by Jorge Luis Borges today. In addition to being beautifully written, it holds some interesting lessons for social scientists (sociology, economics, poli sci...) or people generally interested in explaining the world.

The story is a fictional, first-person account of meeting a young man who, after being paralyzed, could remember each detail of every moment in his life. He created a new numerical system, where each distinct number could be identified with an image or an identity [aside: this is the actual memory technique used by competitive memory superstars; recent book Moonwalking with Einstein covers the topic in a very accessible fashion]. Unfortunately, as that numerical system abandoned all forms of abstraction, it was incoherent to any other person.

Toward the end of the story, this passage particularly stuck with me. 

It was very difficult for him to sleep. To sleep is to be abstracted from the world; Funes, on his back in his cot, in the shadows, imagined every crevice and every molding of the various houses which surrounded him. (I repeat, the least important of his recollections was more minutely precise and more lively than our perception of a physical pleasure or a physical torment)...
Without effort, he had learned English, French, Portuguese, Latin. I suspect, nevertheless, that he was not very capable of thought. To think is to forget a difference, to generalize, to abstract. In the overly replete world of Funes there were nothing but details, almost contiguous details...

The bolded passage is a worthy lesson for anyone trying to model social interactions, mathematically or otherwise. A common complaint about economics is that it oversimplifies, reducing complex human motivations to "rationality" or "selfishness." However, humans are so complicated that some reduction is necessary in order to think about their behavior clearly.

A map with a one-to-one scale is simply the terrain; a model which includes every aspect of human psychology is just as incomprehensible as having no model at all. Providing guidance or prediction means abandoning detail and abstracting at one level or another. The relevant question is which details to ignore and which to include, and in this regard economics (especially when informed by insights from psychology) performs pretty well.

Funes reminds me of post-modern theorists (and other critics of social science) who demand ever more detail and less "reductive" ways of understanding human behavior. Laudable goals indeed, but by implication, theory becomes incoherent and the theorist cursed to never sleep.

Scarcity and Benevolence

I'm reading Democracy and Decision: The Pure Theory of Electoral Preference. Brennan and Lomasky have a very pithy reply to those who make prescriptive claims based on ethical standards:

...in deriving claims about what states of the world are feasible, one must take account of the scarcity not merely of the standard resources - time, ingenuity, and so on - but also of human benevolence and individual ethical sensibility. As Dennis Robertson remarks, one of the chief roles of the economist is to offer a warning bark whenever someone proposes a policy arrangement that demands much in the way of the scarce human resource "love." On this view, an ethical theory of social phenomena that fails to take adequate account of how people actually behave is at best irrelevant to real-world decision making and at worst deeply misleading. (5)

In other words, policy analysis (and related ethical claims) should deal with the world as it is, not as we would wish it to be. Nearly any human problem could be attributed to a lack of love for other people, but simply exhorting others to be more "loving" is not likely to fix those problems.

Subsidize Courtesy?

By "courtesy" I mean making others feel comfortable during interactions they have with you. This has some overlap but is still distinct from "good manners", which to my mind are more formalized and less situation dependent; manners can occasionally be discourteous (e.g. griping at someone for using the wrong fork at dinner).

In my personal life I consider myself a very courteous person. Why do I bother? Two reasons, the first being social convention and the second being economic reasoning. The cost of courtesy is basically zero, and as anyone who's been through microeconomics will tell you, when a good costs nothing you consume as much of it as you can.

So why are some people discourteous? Either they take enjoyment from others' discomfort, or face some cost greater than zero for exercising courtesy. Both reasons seem a bit silly, but the continued existence of discourtesy is pretty good evidence that both do matter in the real world.

Courtesy helps the courteous person (I enjoy social interactions more, and get what I want with less frictions from other people) but it also has obvious positive externalities: both for the person I am courteous towards, and maybe society in general when people are happier and have less strife. Imagine how productivity would increase if everyone in the workplace was kind and genial (but still firm and resolute) towards each other...

With its zero cost and social benefit, there's probably a lot less courtesy in the world than most people want. So why not have the government subsidize courtesy? Maybe each person could have a Yelp score, and strangers could rate their day-to-day demeanor, then collect a check (or a "tax break" if its more palatable) from the state governor on a monthly basis.

This probably seems a bit ridiculous, and I don't actually think courtesy deserves a subsidy. However, the case laid out above is just as (if not more) solid than many of the arguments made for other goods and services the government does decide to subsidize (education, medical care, the fine arts...).

Why does courtesy fail the laugh test and these other subsidies do not?

Social pressure and personal embarrassment are often enough to motivate courteous behavior ("virtue is its own reward"), but arguably those same forces could spur private contribution toward other public goods.

Or, maybe courtesy is just such a personal decision that the government has no capacity to control it -- just like smoking, littering, and bottle recycling -- right?

If someone can give a clear standard which divides courtesy from these popular "public goods" I'd be curious to hear it.

Point Estimates and Division of Labor vs. Technocracy

In his recent book, Public Policy in an Uncertain World, Charles F. Manski notes

Modern democratic societies have created an institutional separation between policy analysis and decisions, with professional analysts reporting findings to representative governments. Separation of the tasks of analysis and decision-making, the former aiming to inform the latter, appears advantageous from the perspective of division of labor...
However, the current practice of policy analysis does not serve decision makers well. The problem is that consumers of policy analysis cannot trust the producers...I recommend that journalists reporting on policy analysis should assess whether and how researchers express uncertainty about their findings, and should be deeply skeptical of studies that assert certitude. That caution and advice extend to all readers of policy analysis. (pp. 173-174)

The rest of the book is quite good, and makes the case for expressing social science results in confidence intervals rather than point estimates, which give (at best) a feeling of false precision. An example of this methodology in practice can be found in Manski's paper on recidivism with Daniel Nagin (1998).

Returning to the quote above, an implication seems to be that technocracy (having the analysts make the decisions in addition to doing the analysis) might be more desirable than representative democracy. It would foster more accountability on both ends - politicians can't blame analysts for policy mistakes after the fact, and vice-versa - and I'd speculate that the benefits of division of labor in politics are a bit overstated.

However, it's hard to imagine the personality types represented by statisticians and actuaries as running successful political campaigns, so division of labor is what we've got, like it or not. Manski's caution about the false precision of point estimates might be a step in the right direction though. For example, the CBO would be a lot more credible if they gave estimates in confidence intervals, instead of specific estimates that are consistently wrong.

Grounding "Methodological Individualism"

This topic came up indirectly today in conversation about communitarian ethics versus classical liberalism.

Classical liberals usually envision a human subject who can choose who and what to associate with, joining or leaving various communities depending on which offers the best "bundle" of services. This search by individuals, finding their private optimums, is supposed to induce efficiency between providers of services who want to attract as many users as possible. The result, classical liberals say, is a spontaneously generated system of social goods which emerge from individuals' entry and exit into various associations.

The communitarian critique (represented by Habermas, Barber, etc.) is to say that people are social animals; we develop our selves through interaction with others. Classical liberals, according to this account, underrate the role of community which can both enrich life beyond what mere individualism can offer, and also restrict the free exercise of choice which classical liberal social theory depends on.

The classical liberal retort to this is "methodological individualism." It is simply a truism that at the core, all choices must be made by individuals so it only makes sense to ground political/philosophical views on individual choices. A community does not exist beyond the coordinated actions of its members, so to speak of "collective preferences", or similar, is just the fallacy of composition writ large.

Methodological individualism is the basis for most good, mainstream economics and from a practical standpoint I think it makes a lot of sense. Talking about the "good of the community" without addressing the incentives of its members invites fuzzy-headed thinking and bad policies which sound good on paper but never work out in the real world. In practice, most people's everyday political heuristics operate on some notion of communal good, so a strong counter-dose of methodological individualism can often lead to improved conclusions.

But, from a philosophical perspective, it's not clear that this fully resolves the communitarian critique of classical liberalism. Indeed, how can we say that methodologically individualist stance is an objective one, derived by us wise political economists with the uncommon privilege of standing outside all the social structures which block the view of others? To put it differently, if we are always already situated within social roles (to borrow some annoying post-modern rhetoric) then perhaps the classical liberal stance simply emerges from growing up in a society where classically liberal and individualistic values are commonly accepted.

Of course, the communitarian stance falls prey to this as well, because Habermas et al are also socially situated in their formulation of communitarian ethics. They have no privileged "eye from above" stance either. At this point we spiral into a long infinite regress of navel-staring and the entire discussion of political economy is put on hold, perhaps indefinitely.

Is there any grounding for methodological individualism, aside from the tautology that "only the individual chooses for the individual"? I can see two possible resolutions for this dispute.

1. Look to results (pragmatism). Which stance works better? From the historical record it appears that individualist societies have grown and prospered much more so than others.

However, there are no pure cases in the real world, and a dedicated communitarian could probably look for the social roots of Western historical success, and/or piously lament the crisis that overly-individualized society has brought upon us today.

2. Dismiss the search for grounding entirely. It may be that no code of social ethics can claim a solid foundation, and all philosophy is built upon chutzpah, pulling itself up by its bootstraps, and assuming its own foundational values into being. In this case we are left with large mental edifices that have been built upon sand, but appear to be functioning pretty much okay, so maybe it's better just not to worry about these grounding questions at all.

In an earlier time this grounding issue could be resolved with an appeal to God, divinely inspired creation, innate human dignity, and so on. But, in an era of secularized social science, competing social perspectives seem to be engaged in an endless bootstrap-pulling competition.

Future of the Welfare State -- is there one?

This was the subject of a talk I attended earlier this week. Loosely titled "The Future of the Welfare State", it featured two discussants to address the funding and politics of the redistributive state. I'll give the highlights here for those who weren't lucky enough to attend, along with some of my own thoughts along the way.

Professors Kent Weaver and Kimberly Morgan speaking at Mason Hall.

The first to speak was Professor Kimberly Morgan of George Washington University, whose presentation was titled "The American Welfare State: How We Spend So Much Yet Achieve so Little". A large part of her talk was dedicated to clearing up factual misunderstandings about the welfare state.

Some figures she cites that are worth knowing:

• Public Social Spending in United States is ~16% of GDP (2007). Closest comparable countries: Australia, Ireland. Largest in this category is France at 28%.
• In net social expenditure as % of GDP we're in 5th place, next to Sweden and the UK.

Net social expenditure is defined as public social spending, plus tax breaks for social purposes, plus publicly mandated private spending (not done very much in the U.S.) plus voluntary private spending (e.g. employer provided health insurance), minus taxes levied on benefits.

An implication of this net social expenditure measure is that the U.S. implicitly subsidizes many groups through the tax code, e.g. the Earned Income Tax Credit (EITC).

In the private voluntary spending category the U.S. is #1, at above 10% of GDP. Next in line is the Netherlands at ~6%. A big part of this is made up of employment-related benefits; U.S. employers spend about $15,000 per employee in health care.

Comment: I see the logic of including private health care spending as "net social expenditure" because it all goes toward public well-being in some fashion, but I object somewhat to lumping this in with government money as if private employer spending is also a variable of choice for public policy. Admittedly, the entire provision of health care by employers is a rather silly relic of wage controls passed during WW2 which has become the "new normal", but it is still distinct from government provided services and should not be treated as a dollar-for-dollar substitute.

• In average indirect tax rates, the highest is Denmark at 26% (followed closely by other Northern European countries) lowest is the U.S. at 4%. This is largely money taken in consumption taxes, e.g. VAT. The upshot is that while the U.S. spends less in benefits, we also "claw back" less of that money than Europe does.

Startling fact: Once you count up all these different social spending categories the US is roughly equal with Sweden as % of GDP! Sweden is famous for its generous welfare state, but voluntary spending in the U.S. is what makes up the big difference.

This begs the question, if we spend so much why are the outcomes so unsatisfactory? Dr. Weaver identifies the biggest cause as how those benefits are distributed; namely, they aren't progressive enough.

• 18% of the U.S. economy goes to health care. Next is the Netherlands, at 12%. In spite of this the American health care system isn't very good in many respects: we're strong on technology, weak on primary care, and nearly worst on "preventable mortality" among developed nations.
• Distribution of entitlement spending in the U.S by income: the middle 60% gets 58%, bottom 20% gets 32% (from the Center on Budget and Policy Priorities).
• Distribution of "tax expenditures": Top 20% gets 66%, bottom 20% gets about 4%, the rest goes to the middle.

Comment: describing a tax break as an "expenditure" seems questionable to me. Dr. Morgan defends this by saying that this money is "given" (i.e. not taken) to promote various social goals, so we should be observant of which sectors of society are given back to the most. To me, this distribution of "tax expenditures" seems inherently obvious: the top 20% pays the lions share of federal income taxes, so obviously there's just more to give back in the form of deductions. It's simple arithmetic.

Dedicating more tax expenditures towards the bottom 20%, who pay little if any income taxes, basically implies a negative income tax ala Milton Friedman. That's a worthy proposal to make and can be debated on its own merits, but the tinge of class warfare in this framing seems amiss.

The next to speak was Kent Weaver, Georgetown professor of public policy. His talk was entitled "Future of the Americaan Welfare State: Constraints and Options". I took less detailed notes during this section; partially because the coffee was wearing off, and partially because he made his points more often with editorial cartoons than charts or graphs, which I personally find less engaging.

He noted the demographic change occurring in America: there are increasingly less workers per retiree, a problem that is even more pressing for Japan and much of Europe. Politicians are also caught in a trap: they can only keep promising high benefits when economy is increasing.

The incentive for politicians becomes focused on NOT punishing groups rather than finding ways to benefit them. This is an interesting point, and one that could be explored in more detail through public choice economics.

Overall both speakers were good, although I enjoyed Dr. Morgan's half of the talk much more. The facts she presented were useful regardless of which ideological persuasion you came from, while Dr. Weaver was more focused on how to expand and continue the welfare state, not necessarily just reform it to accomplish its goals at the same (or lower) cost.

What I took away from their presentation is that, regardless of whether you think government redistribution is a good idea or not, the current way we're doing it in the U.S. is far from optimal. I think this provides some common ground for liberals and conservatives on entitlement reform: if the goal is to provide a social safety net without going toward cradle-to-grave coverage, policies need to be better tailored to accomplish that goal instead of the hodge-podge we have now. Of course, how to get there from here is the real struggle.

Development of the Cayman Islands as a Banking Center -- paper commentary

Tomorrow at George Mason's Philosophy Politics and Economics (PPE) seminar series, Prof. Andrew Morriss of U. Alabama law school will be coming to present his paper, "Creating Cayman as an Offshore Financial Center: Structure & Strategy since 1960", written with Tony Freyer (also of University of Alabama). The paper is still in a draft stage but quite good, although unfortunately, not easily available online (yet). These are my comments upon reading it.

From the paper, I understood there to be three main causal factors which led to financial market development on the Cayman Islands:

1. Relative independence granted by Britain which allowed institutional experimentation

2. As a seafaring nation there was limited ability to tax the locals directly. Instead they used indirect taxes like import duties and stamps, which stalled the growth of redistributive government and encouraged flexible policy

3. Historical timing. In the 1950s better transport and communication technology were becoming available, and the government encouraged access through air/sea travel. This allowed the banking sector to flourish even though it was far from population centers in Europe, which might not have been possible 50-100 years earlier.

One question this invites: Why not Jamaica? The two islands were administratively linked for many years, but Jamaica took a much different development path. The 1960 Companies Law is some explanation because it encouraged company registration in Cayman rather than Jamaica. The authors also note that Jamaica had a more racially-divisive independence movement while Cayman had been racially mixed for a long time, but this "shock" in the 1960s seems inadequate to explain their full divergence.

I'd argue that the Cayman Islands went into finance because they were so small. Lacking much land, population, or other resources besides access to the ocean, Cayman had few other choices but to adopt capital-intensive industries. Jamaica, on the other hand, has more arable land and also tourism revenue which made banking less of a necessity. I'm sure there's more richness that could be added to this story by someone with more expertise in the region.

I thought the section on development of the medical malpractice insurance market in the 1970s was interesting. While the authors use "crisis" to describe the U.S. insurance market of the 1970s there are several journal articles from around that time which contest that framing, and arguably the insurance market in the 1980s was more crisis-like than the 1970s. Malpractice insurance (especially for non-medical professions, like lawyers) was offered primarily by Lloyd's of London or other "boutique" insurers prior to the 1970s, so nearly any increase in tort claims might be classed a "crisis." I can see why a market opened for more foreign providers as demand for insurance increased, but I don't get a full story from the paper as to why the Cayman insurance sector benefited dis-proportionally (if indeed, they did).

Aside from this, the level of detail and description regarding the Cayman Islands as a tax haven, conflicts with the IRS, fighting laundered drug money, and so on is very impressive. Their conclusion, which emphasized the role of stable institutions in allowing Cayman to offer diverse financial services, summed this up nicely. The authors successfully make their point that strong institutions, rather than shady practices or corruption, were the foundation of the Cayman Islands' success. I just wish there had been more of a comparative element which could have contrasted Cayman against some similar nation to show more concretely how those constitutional differences mattered.

Review of "The Science of Success" by Charles G. Koch

I picked up a copy of The Science of Success: How Market-Based Management Built the World's Largest Private Company. It's a quick read, at under 200 pages, and I'll post a brief synopsis of my thoughts on the book here.

Science of Success starts with an overview of the history of Koch Industries, its early development and business moves, and then quickly moves into the "meat" of the book which is a discussion of market-based management (MBM). Inspired by the Misesian tradition in Austrian Economics, MBM has five aspects:

1. Vision - locating where the greatest value can be generated by the organization
2. Virtue and Talents - getting the right people in the right places
3. Knowledge processes - recording and applying relevant information
4. Decision rights - empowering and holding decision-makers accountable
5. Incentives - giving rewards based on value-creation

(paraphrased from a list on p.26).

The fact that such a book can be written I think illustrates a strong point for Austrian Economics vis-a-vis its competitors. If one believed strongly in the Neo-Classical paradigm, with a Walrasian auctioneer who engineered prices to achieve constant social equilibrium, what role would there even be for effective management or entrepreneurship? Neo-Classical economics does have some powerful tools for predicting or forecasting social behavior, but it's not particularly useful in a non-academic setting (at least in its most common academic applications). Of course, if Koch Industries had followed the most extreme strands of Austrian thinking which are hostile to measurement and quantification, it's also unlikely they would have been so successful. However, the Austrian emphasis on entrepreneurship and discovery does make an interesting kernel for a management strategy.

As a handbook on management, The Science of Success has many aspects that you would expect: examples of successful and failed corporate experiments, dealing with people, and product development within a competitive market. I found the parts on how to work within a regulated market to be particularly interesting because those seem like areas that an Austrian economic approach, which appeals strongly to free markets, might struggle. I just wish the section on "Practicing MBM in a Political World" had been a big longer and more detailed.

In many ways, the book is directed more toward helping people understand the philosophy which has driven Koch Industries, rather than necessarily showing how to implement it in other contexts. As described in the introduction, The Science of Success is primarily aimed at current or future Koch Industries employees who want to know how the company thinks and operates, and only secondarily toward a more general audience. So, if you're the CEO of a young start-up company this should still probably be on your reading list, but maybe not as the first title.

In addition to its core Austrian influences, other management thinkers are also featured in The Science of Success. Michael Polanyi is quoted several times, and W. Edwards Deming's idea of continuous improvement is also a driving influence. While some might read it as an ideological screed, The Science of Success also sticks to mainstream best practices. Many of the ideas expressed in Austrian economic jargon could probably also be found in other management handbooks, just phrased differently.

I do have some small technical criticisms. A chart on p. 27 compares a $1000 value of Koch Industries growing since 1960 vs. $1000 in the S&P 500 over that time -- the divergence is notable. However, that difference is somewhat driven by survivorship bias; there were many small natural resource companies an investor might have chosen from in 1960, but many of them are now defunct, and their former CEOs have no reason to write a book on management. Similarly, entries in the S&P 500 have changed over that time, always reflecting relatively "mature" companies which have already reached a stable size and stopped growing quickly. If you were to pick, say, Microsoft or Xerox from its inception and track it against the S&P 500 over that same time period you'd see the same sharp returns curve, but that's because you're comparing a company known in hindsight to have been successful (information not available to investors at the time) against an index representing a safe return. It's not a fair comparison.

Overall, The Science of Success is worth reading because it lays out the intellectual driving forces of Koch Industries. Given the prominent role that the Koch brothers have taken in the world of ideas regarding libertarian economics and public policy, this is useful background knowledge for anyone engaged in those discussions. To keep a semi-balanced perspective, perhaps I'll pick up some of the authorial works of George Soros next...