China promises to pollute as much as possible

Obama has just signed a climate change deal with China. Essentially,

  • The US agrees to cut its emissions by 30% by 2030 from the 2005 baseline.
  • China agrees to have its emissions peak in 2030, and to have 20% of energy come from non-fossil fuels.

The US target is basically just what the EPA is mandating anyway under the Clean Air Act. US emissions have fallen substantially since 2005, largely because unconventional natural gas has lead to massive coal-to-gas switching, which is a much cleaner fuel. There would be even more switching if states didn’t impose such restrictions on fracking. Ironically, many environmentalists oppose the technologies that have done the most to reduce US emissions: nuclear and fracking.

But what’s strange here is the Chinese side of the deal. This seems spectacularly badly designed. It’s basically a cap on all post-2030 emissions at the 2030 level. Which means they’re incentivised to pollute as much as possible in 2030, to give themselves a more generous cap. It literally encourages them to stockpile coal so they can burn many years worth of production in 2030. No need to build coal plants – you don’t need to generate electricity with this burning – this is just pollution for the sake of pollution.

Now, probably China won’t just pile up resources and destroy them for the sake of destruction. But there are many more subtle ways they could take advantage of this, like bringing forward any planned coal plants, so ones scheduled to start in 2031 instead start in 2029, or delaying the implementation of emissions-reducing technologies till 2031.

Given the potential for this agreement to be actively harmful, maybe we should be grateful it’s probably purely symbolic.

Advertisements

How to Pitch a Growth Stock – Cognitive Bias Edition

Growth investors focus on trying to pick the companies that will grow rapidly for many years to come, hoping to be rewarded by a consummately increasing share price. This can be anywhere from Venture Capitalists investing in tiny startups to enormous mutual funds betting on whether Twitter will continue to put up such strong growth. These companies tend to have high share prices compared to their current level of profitability, but have a good story, and will have rapidly grown in the past.

There are certain strategies that people use when trying to persuade someone to invest in a growth stock. This could be the startup team pitching to a VC fund or an analyst in a hedge fund pitching to his portfolio manager. I have more experience with the latter, so this will focus on companies with market caps above $100m.

Many of these strategies are intellectuality illegitimate. As such, I intend this as a sort of ‘Defense Against the Dark Arts’ – how to spot people using these rhetorical strategies in the wild. Perhaps this will also help people employee these strategies – if so I apologize to the world. I take the virtue of silence seriously – but in this specific instance I think sunlight is the best disinfectant – and hopefully it will make for an interesting blog post.

Choosing a growth stock to pitch

The first step is to choose the right stock. Pick something which has seen strong increases in share price over the last few years. A relatively smooth glide path up is best – don’t pick something that rose 20% in one day and has done nothing else. The goal here is to use the halo effect1 to make people confuse the historical share price movements with the company itself – to make it seem like the company itself actually has the property of steadily growing, rather than this just being a property of history of the market’s valuation of this stock.

However, it can be a good idea to pick something which has very recently seen a sharp fall in the share price. This way, your PM won’t feel like they’ve “missed it” – they’ve got another chance to get in. Regret Avoidance is a powerful effect, and you save them from this. Plus, the recent sharp fall means they’re safe from being the guy who bought in at the peak. That guy will look very stupid, so they’re happy to be safe from his fate.

Of course, you need an explanation for why these have happened. The steady rise is easy – the company is also steadily growing. The sharp fall is harder – people don’t want to invest in things that fall! – but there are some easy explanations on hand. ‘Hedge fund de-levering’ is always available as an excuse, and with any luck will act as a semantic stop-sign.

So far this has actually been pretty intellectually respectable. Or at least epistemically lucky – the first requirement, for steady share price growth, probably means the stock has strong momentum, which has historically been a strong predictor of returns (albeit with high kurtosis and negative skew, so beware!). The recent share price fall means the stocks will do well on short-term reversal, which has also historically been a good predictor. The next steps, however, are more dubious.

Total Addressable Market and the Conservation of Conservativeness

Having selected your company, the first step is to work out what the Total Addressable Market (TAM) for your stock is. Is it a household product? Take the number of households in America and multiply by the frequency of purchase. Is it a car? Look at the total number of cars. A better type of steel? Look at total US steel consumption. The key is to get a really really big number. If the number is insufficiently big, just look at a larger category of which the true market is a subset.

Next, multiply that number by their expected market share. As the company has been growing rapidly, it’s probably been expanding its market share of its current niche. So say you assume they’ll keep their current market share even as that niche grows into a major market. This assumption is conservative, you’ll say, because actually they have been growing their market share. This number should be reassuringly small – say 1%. It’s small size will help reassure people that you’re being conservative. If you want increase the total market they’ll eventually control, scope insensitivity means its easier to increase the TAM size than their market share. It’s obvious that 20% market share is a much more aggressive assumption than 2% (especially if their current market share is 2%), but not nearly so clear than $100000000000 is a more aggressive TAM estimate than $10000000000 – especially if you only present one of the numbers.

Next, assume a profit margin. If their current profit margin is high, just say you’ll conservatively assume no economies of scale. If their current margin is low, or they make no profits, just compare to vaguely similar companies and go for a slightly lower number. If ‘comparable’ mature companies have a 30% margin, say 20%. This sounds very conservative, but actually only reduces their profits by 33%.

Finally, assume a valuation multiple. The company is currently trading on a very high multiple, because the market is expecting rapid growth – maybe 30x earnings, or maybe 500x if you’re Amazon. So simply say you assume they’ll get a market multiple. Going from a 30x multiple to a market 15x multiple will cost you 50% of the valuation – but gain you a lot of apparent conservativeness.

The key principle here is the conservation of conservativeness. You want an estimate for them that is both very large and sounds conservative. To do this, you take advantage of scope insensitivity and arbitrage between the TAM stage and the company-specific stage. By making the company-specific stages (market share, profit margin, valuation) sufficiently conservative sounding, you can get away with an aggressive TAM estimate while keeping the whole thing sounding conservative. Scope-insensitivity means you can increase the TAM estimate at a lower cost of conservativeness than you can the company-specific elements, so there are gains from trade.

So once you’ve multiplied your TAM, market share, profit margin and valuation, you come up with an estimate for what this company could be worth in the future. However, you now deny that this is an estimate. Instead, it’s just an idea of the size of the market – you don’t actually expect they’ll reach it. This explicit denial protects you against any accusations of over-optimism, but you’ve successfully primed your audience on a really high number. If market sentiment is a battle between greed and fear, you’ve helped the greed side.

And a crucial subtlety – that valuation that you didn’t make is what the stock might be worth in the future. Because of the time value of money, you would need to discount that back to get to a current valuation. Since it credibly might take 10 years for the market to mature, even with a moderate 10% discount rate your valuation should really take a 61% hit. But by denying it was a valuation, you’ve avoided this step.

Downside Protection

The next step is to argue that the stock has “limited downside” or “downside protection”. This will reassure your audience that even if everything goes wrong, they won’t be fired. You’re trying to quieten their tendency towards fear, so that greed may reign.

Your goal here is to come up with a plausible story for why the worst-case scenario is 10% downside. 10% is just high enough that it sounds vaguely plausible, but low enough that it sounds reassuring.

There are a variety of ways of doing this. One is to name some assumptions you’re making, reduce them slightly, and claim that is the worst case. If the worst case looks pretty bad, just increase your original assumptions, so the haircut versions are higher.

Another is to look at some recent M&A in the sector, pick the most expensive deals, and argue that if the share price fell at all they’d be acquired on that valuation. There is always some expensive M&A going on, so this excuse is always available.

Finally, don’t forget to add that if the shares fell by this much, you’d think this represented a buying opportunity. This is totally misleading – the fact that the shares might be lower in the future is an argument for buying them then, not now, but it sounds level-headed and responsible. It also helps re-direct attention back to the optimistic forecasts of TAMs.

The actual valuation

Since you denied the TAM was actually a valuation, and the downside estimate was definitely the worst case scenario, you haven’t technically actually done any valuation just yet. Since that is ostensibly what you’re doing, you should actually have a go at estimating fair value. Doing all this has probably taken a long time so far, so you won’t have much time left for this stage.

There is an easy way to do it though. Project a high rate of revenue and earnings growth for the next 2-3 years. Place a multiple on the 3-year-out numbers that is lower than the current multiple. Discount that value back, using a high discount rate, to arrive at a share price 30% above the current level.

Fending off criticisms

Since you know more about the stock, you’ll probably be able to come up with a counterargument to any specific concern they mention, and they won’t be able to judge the issue. Any major concerns you have you can simply omit to mention.

If people ask you about execution risk, agree that it’s a risk, but then explain that’s why you used such a high discount rate. This suffices to rebut their attack. This is totally illegitimate. The discount rate (and implicit Equity Risk Premium) compensates investors for the variance of expected future earnings/dividends. It does not compensate you for lower expected earnings/dividends. Their concern is for the latter – your model implicitly assumes everything goes perfectly, whereas a true calculation of expected profits would include probabilities of lower performance. Your attacker will almost certainly not appreciate this fact. If by some misfortune they do, explain in an exasperated manner that you’ve already covered the downside case, and then change the subject.

As a last resort, simply utter the sacred words: “high risk, high reward.” People treat this utterance like magic – we want high rewards! Of course, it’s also nonsense. The phrase should be “high risk, high expected reward.” There is no guarantee you will get the reward! That is why it is ‘high risk’. But as people intuitive understanding of risk is terrible, you can safely abuse the phrase, just like everyone else.

Conclusion

So there you have it – how to pitch a growth stock. Or hopefully, how to spot the intellectually dishonest manoeuvres that are frequently used in growth investing.


  1. At least I think Halo effect is the right one. It also seems to have something to do with the Fundamental Attribution Error 

The Future of Socialism is Privatizing the Atmosphere

Scott recently wrote a quite interesting review of John Roemer’s ‘Future for Socialism’. It sounds rather like Schumpeter’s Capitalism, Socialism and Democracy. Funnily enough his vision for socialism also reminds me of Margaret Thatcher’s plan for property-owning democracy. I haven’t read the underling book, so I can’t comment on that, but I can talk about Scott’s writing. More importantly, it provides an excellent excuse to talk about how to save the environment by privatizing the atmosphere.

Central planning could never work, so a socialist economy doesn’t need it. Bosses and managers seem to be doing a good job keeping their firms profitable, so they can all keep their jobs under socialism. Everyone has different skills, so clearly in a truly socialist system they deserve different wages, in fact whatever wage the market will bear.

…you give everyone an equal amount of these stocks. When the corporations make money, they pay them out in the form of stock dividends, which go to the people/stockholders. So every year I get a check in the mail representing my one-three-hundred-millionth-part share of all the profits made by all the corporations in the United States.

We’ll assume that when Scott says ‘Stock Dividend’, he actually means ‘non-stock dividend.’ A stock dividend is when a company gives extra shares to its existing shareholders. This increases the number of shares outstanding, but has no real economic impact. What Scott presumably means is ‘cash dividend’, which is when a company gives cash to all its shareholders.

However, this immediately opens up a problem with the next part.

Roemer proposes a law that stocks cannot be sold for money, only coupons and other stocks. Every citizen is given an equal number of coupons at birth, trades them for stocks later on, and then trades those stocks for other stocks. This allows smart citizens to invest wisely, and allows a sort of “stock market” that sends the correct signals (this business’s stock price is decreasing so maybe they’re doing something wrong) but doesn’t allow stock accumulation by wealthy capitalists.

While I applaud Roemer’s attempt to make use of the valuable signals sent by prices, his plan for preventing people from selling their shares won’t work. If such a policy was instigated, there would probably be strong demand from people for a way to turn their shares into cash. They’d even be willing to accept a discount for the sake of the liquidity cash offers. So some companies would sell all their assets and pay out all the money as one massive liquidation dividend. By announcing this in advance, the company would basically become a way to turn your shares into cash – just swap your other shares for its shares, and then wait for the single massive dividend.

In this system, businesses would raise funds not by selling stock but by seeking loans from banks.

This is where it really gets crazy. Earlier on Scott said that companies would pay out all their profits as dividends. So they can’t issue new equity, and they can’t retain the profits they’ve earned: companies would eventually become 100% debt financed. As soon as they hit the slightest downturn, without a buffer of equity to absorb losses, they would all go bankrupt. And bring the banks down with them. Then you have zero companies, shareholders would envy those who got their money out before the end, and the living would envy the dead.

So perhaps we’ll lighten the requirement that companies have to pay out all their profits. Companies that routinely raise new equity will be in trouble, like tech companies, but lets assume they solve that problem. Utilities also rely on continued equity issuance, so we won’t be able to charge our devises anyway.

More seriously, this would present massive problems for new companies. Or rather, it would prevent there being any new companies. The way you found a company is by investing some money and becoming the owner of a startup – effectively, the startup sells stock to you. Without this, there’s no way to found a new company. So we have a finite number of companies, that occasionally go bankrupt, take each other over, or liquidate themselves. These companies own all the factors of production, leading to a less and less competitive economy, dominated by a couple of few firms, with absolutely no fear of new entrants shaking up their cosy oligopolies.

So there are some problems with Roemer’s ideas. In fairness to Scott, he spots a lot of other problems himself, and he doesn’t even have an economics background. In fairness to Roemer, perhaps Scott misrepresented him. Lets just say that Roemer-as-paraphrased-by-Scott’s plan has some serious disadvantages.

However, it did make me think of an interesting idea I had a while ago. Here is an way of using joint-stock corporations to solve collective action problems.

How to solve the problem of pollution by privatizing the atmosphere.

At the moment, people are incentivized to over-pollute. If my factory releases dangerous emissions, I get much of the benefit, in the form of profits from selling my product (along with my customers, employees, suppliers etc.) I pay only a fraction of the costs though – most of the pollution effects other people. Since I gain much of the benefit, and little of the cost, I tend to pollute too much.

The problem here is one of negative externalities. Equivalently, it the tragedy of the commons. And what is the solution to the tragedy of the commons? Privatization. If one person owns the field, they have the right incentive to preserve its value.

Similarly, we could privatize the atmosphere. People who wanted to use the air (say, by breathing, or burning fuel) would be charged a fee. This would cause them to internalize the external cost, and restore efficiency to the market for pollution.

Of course, this would be rather difficult to administer. How are we going to charge people for breathing? Do people get charged more for having bigger lungs? If people fall behind on their payments, do we cut off their oxygen? Doing so would plausibly count as theft, as currently they enjoy use-rights to the air.

Fortunately, this can easily be solved. Simply give everyone shares in AeroCorp. Because AeroCorp gets most of its money from coal plants and gasoline companies, it pays a dividend each year well in excess of the breathing price. So everyone’s breaths are just netted against the dividend, and they never have to send any money to AeroCorp. Because polluters now have to pay AeroCorp to emit pollutants, they’re less keen to do so, and the negative externality problem is solved.

We could even have a dual share class system. Every human is given a single A-share at birth. These are non-transferable, dilution-protected, and their purpose is to ensure that everyone can afford to breathe. We also have B-shares. These have the same voting and dividend rights as A-shares, but are transferable and dilutable. These are initially auctioned off in a standard IPO. They money raised will be used to fund AeroCorp’s operating expenses. Trading in these shares would ensure price discovery, efficient capital allocation and allow secondary issuance.

There some problems with this system. For example, the firm would be a monopolist, so would tend to charge polluters too high a fee. As such, society would actually end up underpolluting.

Additionally, we need to ensure the two share classes don’t take advantage of one another. There are probably more A-shares than B-shares, but B-shares will be more closely attended to.

One strategy B-shareholders could use would be to have AeroCorp buyback stock. Ordinarily this would be fine – it would raise the value of A-shares. However, in this instance we’re relying on the dividends paid to B-shares to cover the oxygen charge.

A strategy A-shareholders could use would be to insist on new equity issuance, diluting B-shareholders, then paying out the funds raised as a dividend, thereby benefitting themselves.

These two problems could be solved in an attractively symmetrical fashion by giving the B-shares a veto over buybacks and giving the A-shares a veto over new equity issuance.

GiveWell is not an Index Fund.

Someone1 on facebook recently asked

Can we think of donating to GiveWell- or GWWC-recommended charities as being the philanthropic analogue of investing in an index fund? In the sense that it may be possible in principle to do better, but it’s close to the best among readily available options, and almost everybody who tries to do better will do worse.

I think this analogy overlooks some important points about the underlying structure.

An Index contains all the stocks that satisfy some very broad criteria – for example, the S&P500 is basically “is very large US company”. Index funds invest in all the stocks in a specific index.2 Index funds try to do as well as the index on the whole – no better, no worse.

Active managers, on the other hand, by small subsets of the stocks in the index, and try to beat the index. That is, they try to buy stocks that will do better than the average stock in the index.

Part of the appeal of index funds is that many people think that all stocks have basically equally good prospects, ex ante, due to the Efficient Market Hypothesis.There are good reasons to think that free markets are in general very efficient.

Givewell looks at a large number of charities, and selects a very small number to recommend. In this way they’re much more like an active manager than an index fund. An index fund for charities would mean spreading out your donations between thousands of different charities.

There are important differences between the stock market and the charity market. Virtually any time you think you see an inefficiency in the stock market, you are probably wrong. But this does not apply to the charity market. Effective Altruists habitually and credibly claim that there are many orders of magnitude difference between the expected values of different charities (pdf). And the sorts of arguments about risk and diversification that motivate balanced portfolios of investments simply do not apply to charities; with charities, you should pick the best one and donate everything to it.3

Some perhaps Index Funds are Givewell are similar insomuchas they are both good things. But so is chocolate.

This analogy glosses over important differences, and potentially causes sloppy thinking. Just because two things are good doesn’t mean they are good for the same reasons or in the same way. Effective altruists would do well to understand that markets are not like charities. Others would do well to understand the there is no Efficient Charity Hypothesis!


  1. who will remain anonymous until they request otherwise. 
  2. technically some merely invest in representative sub-baskets, but S&P500 indexes probably buy all the constituents. 
  3. unless you are very rich. 

Population Growth and Innovation

In a world with substantial population growth, the natural tendency is for capacity utilization to rise over time. We built enough machines to produce as much stuff as was demanded by a population of a certain size, but then the population grew, so now demand exceeds supply. This raises prices, and means it is profitable for businesses to invest in new equipment to meet the new demand. At this point, businesses will look at the different types of machine on offer, and go for the one that offers the highest return. If you can design a better machine, that either does more than the old ones at the same price, or does the same for a lower price, firms will buy your machine rather than the older designs.

Without population growth this doesn’t work so well. With no incremental demand, the existing stock of machines is adequate for demand. Ignoring depreciation, firms don’t need any new machines. So to sell your new machine, you need to persuade a business that not only is your new design better than the old ones, it has to be a big enough improvement to justify the entire cost. In the population growth case, your design only had to be a little better, as businesses were in the market for a new machine anyway; now your design has to be good enough to justify getting rid of an old machine!

It seems that the first world, with population growth, is more supportive of innovation than one without population growth.

There is a common argument that a high population level is good for innovation, because it means there will be more scientists, engineers and entrepreneurs whose discoveries can be spread across everyone. The argument I just outlines is different, however, in so much that it relies on the rate of change, not the absolute level.

2014 Shadow FOMC Statement

Firstly I would like to thank the FOMC for allowing the creation of the Shadow FOMC. In these times of controversial monetary policy, it seems only prudent to have a loyal opposition. And with the increased importance of the Shadow Banking System, whose ranks have been swelled by refugees fleeing the forces of Genghis Frank and Khan Dodd, it is vital that said loyal opposition should be a Shadow FOMC.

We intend to issue quarterly statements, coinciding with the releases of the Light FOMC .

Release Date: September 19, 2014.

For immediate release.

Information received since the Shadow Federal Open Market Committee last met, which was never, suggest that the early universe rapidly expanded due to cosmic inflation. However, the expansion was disjointed due to quantum fluctuations. The effects of these imbalances continue to have a major effect on the universe, as they are the cause of all variation, including galaxies, planets, and the fractional reserve banking system. More recently, fear of another kind of inflation lead to mistakenly tight monetary policy during 2008, a key cause of the recession, from which the economy has now partially recovered.  Private Sector GDP is running at a trend rate almost as high as during the Clinton and Bush II bull years, though many analysts miss this due to a mistaken focus on total NGDP, which includes government spending at cost. Industrial production rose. The S&P500, which appeared on the verge of mass bankruptcy, is now at record levels, seeing a 30% rise in 2013. Unemployment, painfully high for a very long time, has started to accelerate downwards. Employment figures, however, have been very disappointing; partly because of demographics, many of the unemployed have simply given up looking for a job. Only recently have we seen the employment/population ratio begin to rise.

More recently, since the Light FOMC met in July incremental data suggests that economic activity is expanding at a moderate pace. On balance, labor market conditions improved somewhat further; however, the employment rate has only improved slightly and wage inflation is muted, suggesting there remains significant underutilization of labor resources. The minimum wage continues to distort labor markets, suggesting a need for higher inflation to reduce its effects. Household spending appears to be rising moderately and business fixed investment is advancing, while the recovery in the housing sector remains slow. Restrained fiscal policy is encouraging true economic growth, although it is confusing accountants who mistakenly assume government spending is by definition useful activity. 10-year breakevens have recently fallen back to 2%, and 5-year breakevens are below 1.7%, slightly below the Light Committee’s longer-run objective. The US government has not issued NGDP-linked bonds, so we lack market implied growth figures, but trailing Nominal Private Sector GDP (NPGDP) growth has been trending around 5%, almost at our 6% target – and substantially closer than the irrelevant NGDP, which is closer to 4%.

If it was acting consistently with its somewhat misguided statutory mandate, the Light FOMC would not have initiated tapering, as low employment and low inflation both suggest a need for lose monetary policy. The Shadow Committee agrees that, with current mediocre policy, economic activity will expand at a moderate pace. The Light Committee thinks that the likelihood of inflation running persistently below 2 percent has diminished somewhat since early this year, which is very strange, given that breakevens have fallen since then. Maybe the Light Committee could not afford a Bloomberg terminal.

The Committee currently judges that there is sufficient underlying strength in the broader economy to support ongoing improvement in labor market conditions, but that this does not excuse irresponsible monetary policy. In light of the hugely reduction in the employment/population ratio since 2006, but improvement in the outlook for labor market conditions since the inception of the current asset purchase program, the Committee decided to maintain the pace of its asset purchases. In October, the Shadow Committee will continue to add to its holdings of agency mortgage-backed securities at a pace of $10 billion per month, and will add to its holdings of longer-term Treasury securities at a pace of $15 billion per month. Both Light and Shadow Committees are maintaining its existing policy of reinvesting principal payments from its holdings of agency debt and agency mortgage-backed securities in agency mortgage-backed securities and of rolling over maturing Treasury securities at auction. The Committee’s sizable and still-increasing holdings of longer-term securities should maintain downward pressure on longer-term interest rates, support mortgage markets, and help to make broader financial conditions more accommodative, which in turn should promote a stronger economic recovery and help to ensure that NPGDP, over time, is at the rate most consistent with the Shadow Committee’s mandate.

The Committee will monitor incoming information on economic and financial developments in coming months as closely as it can be bothered, and will continue its purchases of Treasury and agency mortgage-backed securities, until the outlook for the labor market has improved substantially in a context of NPGDP growth. If incoming information broadly supports the Committee’s expectation of ongoing improvement in labor market conditions and the Committee’s optimistic hope of NPGDP moving back toward its longer-run objective, the Shadow Committee will end its current program of asset purchases at sometime in early 2015. However, asset purchases are not on a preset course, and the Shadow Committee’s decisions about their pace will remain contingent on the Shadow Committee’s outlook for the labor market and broader economy, as well as its assessment of the likely efficacy and costs of such purchases.

To support continued progress toward maximum employment and price stability, the Shadow Committee today reaffirmed its view that a highly accommodative stance of monetary policy remains appropriate, and suggests that the Light Committee might like undertake one, rather than merely talking about it. In determining how long to maintain the current 0 to 1/4 percent target range for the federal funds rate, the Committee will assess progress–both realized and expected–toward its objectives of maximum employment and 6% NPGDP growth. This assessment will take into account a wide range of information, including market indicators, what witty people say in our twitter feed, and patterns we see in clouds. The Committee continues to anticipate, based on its assessment of these factors, that it likely will be appropriate to maintain the current target range for the federal funds rate for a considerable time after the asset purchase program ends, especially if projected NPGDP growth continues to run below the Committee’s 6 percent longer-run goal, and provided that longer-term NPGDP expectations remain well anchored.

When the Committee decides to begin to remove policy accommodation, it will take a balanced approach consistent with its longer-run goals of maximum employment and NPGDP of 6 percent. The Committee currently anticipates that, even after employment and inflation are near mandate-consistent levels, economic conditions may, for some time, warrant keeping the target federal funds rate below levels the Shadow Committee views as more consitant with the fundamental level of human time preference, which implies long-run real rates of around 3-4%. Over the longer term, the Shadow remains concerned not only about technological stagnation but also about the risks from Superintelligent Artificial Intelligence.

Voting for the FOMC monetary policy action were: everyone, because of the Aumann Agreement Theorem.

Castle Consolidation as a raison d’être for the EU

People have given many (usually quite poor) arguments in defense of the EU. Perhaps this is because the EU is actually a quite poor quality institution. However, there is one argument for it that I have never seen considered in the literature: the argument from Castle Consolidation.

Castles are an excellent example of an industry fallen on bad times. Huge amounts of investment were poured into them a long time ago, but demand for their services fell over the centuries, as they were rendered obsolete in their primary market by new market entrants, like gunpowder and compacted earth forts. Regulatory change (the decline of feudalism) also hurt their profits. It’s safe to say that castles are no longer a worthwhile investment. Returns on invested capital1 are low, which is why few private equity funds are building new castles.

There are a great many castles in Europe, all in competition with each other for tourists and film production. What the industry needs to do is consolidate; if it could get down to a smaller number of firms, they could collude to raise prices. Import threat is limited, because although there are some nice ones in the Middle East like Krak des Chevaliers, shipping costs are prohibitively high. Returns are currently so low that they have room to rise substantially before new entrants are attracted to the market. There is little room for substitution because castles are awesome.

English Heritage has already successfully consolidate most of the castles in the UK; what remains is cross-boarder consolidation. That, presumably, is where the EU comes in: as a castle cartel.


  1. If you’d like to learn more, I recommend Damodaran

Average Utilitarianism and Agriculture

This post makes an argument that, if you believe A, you have some reason to believe B. I don’t believe A, but hopefully I have done a good job of mentally modelling the concerns of those who do. Please note that “but A is false” is not a valid response to this post (ex falso quodlibet notwithstanding).

On Agricultural Matters

Suppose you are an average utilitarian, who only cares about the average level of human happiness.1 Suppose further that all crops (wheat, rice, soybeans etc.) are used for human consumption – there are no ethanol or biodiesel industries, for example.

In the short-run, the supply of crops is mainly dependant on the weather. 2014 is looking like a good year for the US crop, as was 2013, while 2012 was bad. US corn production was 29% higher in 2013 than 2012, which was itself 13% lower than 2011. Short-term variations in crop supply are mainly due to weather, but the long run average volume comes down to the acreage planted and the amount farmers invest in raising yields (tractors, GM seeds, fertilisers, etc).

In order to prevent occasional famines, where insufficient crops are produced to feed people, you need to make sure farmers plant and invest enough to ensure that even in bad weather years, there will be enough harvested to feed everyone. Unfortunately this means that in most years, where the weather is not awful, there will be significantly more harvested than is required. Demand for bread is quite inelastic: we need a certain amount to live, but we’re not interested in eating very much more than that. So in years of good harvests, supply would massively exceed demand, and the price of crops would plummet to a low level, as happened this year. As most years do not have exceptionally bad weather, in most years prices will be very low – which will not encourage farmers to plant enough. As such, farmers are likely to under-plant so as to keep expected (average) profitability reasonable, which will ensure famines in years with bad harvests.

One solution is to stockpile grains between years. This is so straightforward it doesn’t warrant further comment.

Another is to make the demand for crops more elastic, so that even in good harvest years there will be sufficient demand. Setting aside moral qualms, in theory the government could do this, for example by buying excess crops to turn into ethanol. However, it is important not to confuse the omniscient, benevolent government planner of economists’ models with actually existing governments. The real-world implementations of such policies, like the US ethanol mandate or the Common Agricultural Policy, have been awful.

Fortuitously, there is a natural mechanism in place that makes the demand for crops elastic; meat consumption. As meat is a luxury on the margin (though some level of consumption seems to have substantial health benefits), demand for meat is significantly more sensitive to price than food in general. And it requires a large amount of grain to make a relatively small amount of meat. So farmers plant and invest enough to supply the demand from both humans and cattle herds in good times; then in years of exceptionally bad harvests, the price of grain rises, so animal husbandry is no longer economic. Farmers slaughter their herds, and the grain they were consuming is now available for human consumption. Even better, there is a short-term massive supply of beef, which can also help make up for the poor harvest. (I guess this is basically a way of storing grain inside cows.)

This reasoning is significantly more persuasive to average utilitarians than total utilitarians. By supporting agricultural investment this system helps prevents famines, which presumably lower average happiness. But it keeps the overall human population lower than it could be. In years of good harvest, the grain that is slowly wasting in storage, or being turned to Ethanol, or being fed to livestock, could instead by directly feeding people, and supporting a higher population, albeit one prone to periodic famines. The total utilitarian would also have to take into account how much pleasure people get from meat consumption, how much displeasure is caused by famines, and how many additional people could be supported on a more vegetarian diet.


  1. I think this is a silly view: it might commit your ethics to massive dependence on unknowable alien populations; it might require you to murder millions or billions of people for being insufficiently happy; it might force you to create really miserable people to ‘dilute’ the effect of sufficiently many even more unhappy people. And perhaps we should be concerned about the welfare of animals too. But suppose.