Showing posts with label tradeoffs. Show all posts
Showing posts with label tradeoffs. Show all posts

Monday, May 17, 2021

Two Methods of Improvement

Let's compare two general methods of improvement: 
  1. Truncating the left tail so as to eliminate the undesired portion of the distribution
  2. Increasing the distribution so as to grow (fatten) the right tail and therefore increase the desired portion of the distribution. 


Both methods have the effect of shifting the mean rightward. But the first is artificial.

Let's explore the first method. People paid primarily for their looks are an example of truncating the left tail. (One might be tempted to say “supermodels”, but that is a particular, specialized subclass of this universe. It is like saying basketball players when we are actually talking about athletes.) They exist within a distribution of attractiveness (subjectively considered as that is the only way) that simply has lopped off most or all of the left side. Some are gorgeous to you; others are gorgeous to me. Some are not so attractive to you while others, perhaps ones you really like, are not so attractive to me. Anyone in particular within this group might be just okay to any random observer. Taking everyone's opinions together as a whole, though, on average gives us an ordered distribution [similar to the theoretical and problematic Keynesian Beauty Contest]. 

When considered from the average observer’s viewpoint, the only thing missing in the distribution are all those who would be below some threshold. In other words the “lowest” (most left) person paid for looks is just an average looking person compared to all of humanity. Because we can’t manufacture attractive people yet, we are forced to use the truncate strategy. 



So the only way to bring about beauty improvement is by leaving out those who are less than some level of beauty (I used eliminated everyone below the average beauty score in the example). So we can get there, but it is artificial--we just left out the less than "beautiful", whatever that actually means in this hypothetical.

Now think about wealth. How do you increase average societal wealth? This is problem from a different realm because unlike beauty where we are currently limited to some degree of diet control, physical fitness training, and plastic surgery we can move wealth around. 

In the case of wealth what is the better path: Minimizing the impact of bad ideas (truncating the left tail via redistribution) or increasing the rewards for good ideas (fattening the right tail)? 

Bailing out bad ideas has moral hazard risks--we are subsidizing bad ideas. When you subsidize something, you get more of it. Taken to the extreme income redistribution is not sustainable. The system collapses in on itself through actual complete resignation (a dead-end Nash equilibrium) or deliberate exit (John Galt). Because of this, we are forced to use the grow the distribution strategy. 


Notice how this distribution is truncated and non normal (there is a minimum at 70 and the distribution has a right skew). No one is below some level of actual wealth (even debtors and prisoners get a meal and a place to sleep). So in some sense I am assuming some of the first strategy--a social safety net of some kind. I wanted to make it more realistically skewed, but time didn't permit. However, we should be careful how easily we succumb to the notion that there are people with true wealth at the far, far reaches of the distribution. Just how rich is Jeff Bezos compared to you or me really, seriously

Growing the distribution has a side benefit of minimizing the impact of bad ideas--a kind of resistance to bad ideas having meaningful, lasting impact. Subsequently the opportunities for good ideas are increased since this method is positive sum (it grows the pie) while the former strategy is zero sum and eventually negative sum if taken too far.

Am I assuming too much? I really don't think so. 

Unfortunately, advocacy for method two is unpopular because of social desirability bias. People don't want to admit that they want the rich to get richer. Or worse yet, they think letting the rich get richer somehow makes us all worse off. 

Sunday, May 16, 2021

Dynamic versus Static Investments

One way to categorize investments would be dynamic assets like stocks and bonds and static assets like commodities and art. 

Investment taxonomy is multidimensional. Along one dimension would be concentration/dispersion diversification. For example owning stock in one company alone as compared to several companies within the same industry as compared to several companies among different industries. Still a greater level of diversification can be achieved as an investment portfolio approaches a share in all companies (e.g., broad-market index funds).

Along another dimension would be type of return claim. A stock investor has a residual claim to the profits that remain after all other claims have been satisfied. That is after all creditors have been paid--all liabilities have been settled. A bond investor (lender/creditor) has a primary claim putting them somewhere higher up the priority list. Keep in mind there are various levels for various types of debt issued. 

Along another dimension would be whether the investment generates cash flows or is dependent upon perpetual new buyers at higher prices for its rate of return. Consider my prior partial list and the explanation behind the distinction

Still another dimension would be how dynamic are the assets one is investing in. Can the asset maintain or gain value in a changing world? This is where I would like to focus today.

In my thinking dynamic assets are essentially investments in ideas with potential cash flows. Static assets are essentially investments in inputs (costs) where the market is continually working against you and high-risk speculations on future tastes and technologies (future desirability & greater fool theory rolled into one speculative bet). 

With static assets you can be right and still be wrong. The opposite is true of dynamic assets to some degree as businesses can change direction

Since static assets have a locked-in nature, they should command a risk premium. Indeed they do but it isn't necessarily sufficient compensation for the added risk of loss they offer. 

The general case is that the more dynamic an asset is the lower its experienced volatility and thus the lower its expected return. But there is more to this story. Volatility is a cruel mistress. It can rob an investor by impairing capital because of the pattern of returns--negative or even just low returns at the beginning of an investment horizon while cash is being pulled out can leave the principle so low that eventual recovery is not possible. Therefore, a static asset with high volatility and high expected return might experience high (negative) volatility, prices going down rapidly, at just the wrong time, early in the investment horizon. Dynamic assets can be the slow and steady that wins the race. 

Volatility's cruelty doesn't end there. It can collapse and vanish as well, but this leaves investors with low expected returns. A static investor needs volatility to justify future returns. 

One should not assume I am saying that one is preferred necessarily to the other. Rather this is one exploration into how assets can be categorized and how to think about investments. Assets along this spectrum fulfill differing objectives with differing opportunity/risk characters. Investing is about tradeoffs.

Consider this stylized linear example: 



Forever people have been trying to eliminate oil: first they were doing so because it was nothing but a nuisance, next they we're doing so because it was getting more and more expensive as more and more of the world's machines ran on it where the solution was to find more and more of it and produce it cheaper and cheaper, and now substitutes are becoming a better and better option. 

To bet on a technology one needs to be compensated for risk with commensurate returns, and because the chance of a given technology profitably working is incredibly small, those need to be exceptionally high potential returns. In that case don’t own oil; own mineral rights. Don’t own proven reserves; own unproven reserves. Don’t own production; own potential production. 

Oil was to the 1920s what cryptocurrency assets might be to the 2020s. Bitcoin, Ethereum, and all other crypto assets are bets on a particular strategy within a particular technology.  So invest with care. These are highly speculative and certainly very static within my classification. Do not mistake the dynamic ability of people and firms with ideas and fluidity to be attributes of the underlying crypto technologies these people and firms may be employing. 

One last example: Diversified real estate is a dynamic asset while concentrated real estate is static. Similarly investment in the rights to a franchise within a geographical area is more dynamic than is the franchise location itself and even more so than the specific land the franchise sits upon. To this end see the picture below and keep looking until you see it:



Wednesday, May 5, 2021

Annuities - A Troubled Solution in Search of a Problem

Years ago I'm sitting in a San Francisco coffee shop with my wife enjoying breakfast. Without trying to or really wanting to we can easily hear the conversation from a close-by table. It was two young couples. Both were well dressed, but one was decidedly more outgoing and charismatic. One might even describe them as smooth.

They were clearly on travelling together. Somehow their conversation turned to topics that drew my attention. It began innocently enough.

"Well, what are your plans?" or so went the inquiry. "Nobody wants to think about this stuff, but it is important." They were clearly talking about someone who wasn't there. 

"It is hard to know what to do."

"Look, we obviously can't know the future. But with this approach at least you have something to show for it..." Turning to her partner a little too on cue, "Remember Grandma’s experience..."

I don't remember too vividly the exact conversation--I honestly wasn't trying to listen.* It was not a simple case of a couple-friend giving friendly advice. This was a sales pitch. And they were selling the other couple on the idea of long-term care insurance, a type of annuity that has very strict terms regarding when it will be paid along with sharp limits on how much and long payment will occur. 

LTC insurance plans are not bad per se. They can work in practice; though they more frequently work in theory. While I didn't know all the relevant facts in this situation, and it was none of my business regardless, the conversation frustrated me. In fact I was offended. Why?

I was offended because they were using emotion to solve a math problem. Well, more precisely they were disguising an emotional pitch as if it were a math problem, pretending it was a math problem, and not doing or even hinting at any math! 

Presumably there would be some assumption-laden work-up presented at some point before signing on the dotted line. Let's charitably assume there was--that all we were witness to was the initial hook. Regardless, I resented both the approach and the fact that it appeared to be working.

It was a learning moment for me. As analytical as I want things to be, the truth is humans are emotion-driven beings. Many of our decisions are based on feelings. We seek social desirability and find comfort in confirmation. 

This is why confident people are charming. Especially it is so when they are selling us something. 

How you say it versus what you say--delivery versus content. They will remember how confident they were in you long after they have forgotten what you actually said. 

I remembered this story as I read this recent piece from Vanguard, Guaranteed Income: A Tricky Trade-Off. From the summary bullet points:
The math is clear. A certain income can leave retirees better prepared for an uncertain lifetime. But retirees’ reluctance to annuitize suggests that the irrevocable decision to exchange liquid wealth for guaranteed income is about more than math.**
It is not too much of an exaggeration to say that there are two types of people in the financial products industry: those who sell annuities and those who detest them. A derogatory but perhaps not unfair way of describing annuities is to say that they are never bought always sold. Another is that the primary beneficiary on a variable annuity is the sales person.

Annuities work extremely well in theory. They are straightforward instruments that spread risk and smooth income. 

In practice they are extremely complicated, notoriously misleading, and very expensive. There are exceptions. The regulations around them have improved the situation some, but I would argue strongly that this is a second-best solution behind simply allowing more competition in the industry in the first place. World-class fine dining in Napa Valley isn't because of world-class restaurant regulation. 

If you're paying attention, you'll have noticed a paradox. I started by showing that people often use emotion to sell a financial solution but then argued that emotion is keeping people from adopting those same financial solutions. But that really isn't a mystery. If people are reluctant to listen to the clear math supporting annuitizing future income, it stands to reason that emotion will be perhaps necessary to get them over the hump. 



*In fact they were so bad at attempting to be discrete that I can only assume we too were part of the sales audience.

**The Vanguard piece points to fear of regret and a strong bequest motive as the major obstacles to annuity adoption. I liked their analysis, but I don't think they sufficiently considered just how few good, honest annuity options there are. Hard to buy what isn't being sold--especially with fair options that do leave bequests. And it is harder and harder to sell them. Whether deserved or not (it is definitely deserved!), annuities have been given a bad name by all the many investment advisors who rail against them. 

Sunday, May 2, 2021

Zoning Laws Suffer From The Fixed Window Fallacy

The Fixed Window Fallacy is an error in reasoning whereby people believe they know or can know what is nice/preferred/optimal. This line of thought is based on unimaginative, linear-thinking and further held back by the Local Maximum Problem

It can be summarized as a thought process that goes: "We know what is best. We/they can afford what is desired (after all, it is usually for our/their own good). Therefore, we should make ourselves/them provide it." 

Both premises are false, and the conclusion is fallacious (non sequitur) as it ignores the critical questions: do we have a right to do this, and can we successfully do this? 
The only constant is change, and it comes in two types. 
  1. Depreciation, which is the natural condition, difficult to counter, and mostly objective.
  2. Appreciation, which is the abnormal condition, difficult to achieve, and highly subjective. 
Attempts to stop depreciation such as zoning laws are never done in a vacuum. They are not single events where good replaces bad, and we move on to the next decision. They are part of economic evolution where decisions made affect trend trajectories with uncertain net outcomes and unpredictable magnitudes. 

Similarly collective action attempts to realize appreciation such as subsidizes for development and master plans are fraught with captured interest risk bringing asymmetric outcomes adverse to the presumed collective goal. In other words the rent-seeking developers and their friends in power do what is good for them and costly for society. For those cases where everyone has the best of intentions, we still have the knowledge problem. When artificial outcomes are engineered by those who do not bear the full risk, bad ideas do not get properly punished and good ideas do not get properly rewarded. 

Back to zoning, trying to stop people from doing things they want to do is prohibition. People and markets work to thwart prohibitions in proportion to how much they desire that which is prohibited. The less morally sound the prohibition, the less compliant are those working against it and those third parties who have no dog in the fight. Fortunately the long-term trend is for less and less prohibition. Unfortunately working against a prohibition is costly as is the administration of a prohibition. 

Whether it is in icky markets (e.g., sex work, recreational drugs deemed illicit, kidney transplants, etc.) or in we-know-better markets (e.g., zoning), an underlying force supporting the prohibition is not in my backyard thinking. In fact I believe NIMBY is the last vestige of prohibition rationalization.