Saturday, April 30, 2022

Choose: Stocks and Bonds or Bitcoin and Cash

Over lunch this past week an interesting hypothetical was posed. Suppose you were offered one of the following, which would you choose: 
  • One million dollars in some initial combination of your choosing between stocks and bonds (fully-indexed, total market coverage), or
  • One million dollars in some initial combination of your choosing between Bitcoin and cash (U.S. dollars).
You will be forced to lock it in for 10 years with no changes to it or any ability to borrow against it. After the 10-year period is up, it is yours free and clear (no taxes either at that point).

Without too much thinking or much hesitation, I chose Bitcoin and cash in a 50/50 combination. My wiser colleagues said with as much or more conviction stocks and bonds--I don't recall their combinations if they stated them. Since I am the investment guy, this raised eyebrows. Maybe I'm just also the gambler. To be sure I caveated my decision with the disclaimer that I might change my mind (my guess was low conviction). To be fair the others did similarly but with perhaps a bit less hesitation (somewhat higher conviction).

In general I would assume that all four of us in this conversation are of very similar financial standing adjusted for our ages (there is about a 30-year spread from youngest to oldest). There is not a right or wrong answer on this question--at least not without a lot more information about each chooser including several underlying assumptions (risk tolerance, liquidity needs, expectations about each person's future goals and paths of life, etc.). I don't wish to get into speculation about that here nor try to evaluate the soundness of any starting position. 

What I am interested in is exploring further how we might frame such a tradeoff. One additional outcome from this exercise is thinking about what assumptions one would make about critical variables and the implications of those assumptions. 

Some people would very appropriately, for themselves, choose an allocation of 100% cash. We could argue about that, but again only by digging deeper into their goals and risk tolerance among other things. "Hey, I'll take free money and I want to know it will be basically there for me at the end of the rainbow (inflation be dammed!)." That is potentially a sensible position, but we could write a book (many books have been written) about what extreme conditions must be in place for that to be rational. Geez, I better stop now or that will be this post . . .

So let's just assume we are debating only the question of which outcome has the best highest expected value after 10 years. I strike "best" because that implies more than just the math problem I want to explore.

We need just a few inputs: 
  • expected returns of stocks, bonds, Bitcoin, and cash (I am assuming we can get some yield on cash rather than thinking of it as money under the mattress.)
  • expected inflation (We are going to look at values in real terms so we don't let the cash option appear better than it actually is--a likely net loser to inflation.)
  • probability of various outcomes (Using a range of expected returns we need to know how likely we think those are. The range is really only important for Bitcoin given its unknown future.)
I am going to use Vanguard's capital market assumptions (CMA) for expected returns of stocks, bonds, and cash as well as inflation. To make these always updating predictions evergreen in this post and because these are publicly available information as linked above, I will also post a picture of these below. Please do see Vanguard's website for more information including appropriate disclaimers. 

I am going to totally make up the expected returns for Bitcoin because 1) my guess is as good as yours and 2) the devil is in the probability and the relative outcome versus the others--my accuracy is nearly immaterial if I am in the ballpark. 

Before you dismiss any of this upon glancing at the inflation prediction (range 1.6% - 2.6%), understand that these are 10-year predictions. I hope they are right given what this implies going forward given currently very high inflation rates, but it can easily be the case even with some persistence of current inflation (8% for a year (not that bad yet) plus 5% for a year plus 8 years at 1.6% would land us at the high range).

Note that I am looking at true total market coverage in stocks (U.S. and all international), thus I will combine the growth rates below in the proportion 55/45 U.S./Int'l. Note also that I am only using U.S. bonds in the model. I generally like some international bonds, but I will make this limiting assumption. Regardless, U.S. versus Int'l bond returns are pretty close as you can see in the details below and at the link.

Enough of that already, let's model this thing.

Here is version 1:


I am putting my thumb on the scale to be optimistic about traditional asset returns (stocks and bonds)  making the low case 25% likely and the high case 75% likely, which serves to put my initial Bitcoin/cash choice at a disadvantage. For Bitcoin I am assuming total collapse in the low-end prediction versus only 25% annual growth in the high-end version. I say only as this isn't "to the moon" although it is very strong growth indeed. Keep in mind that Bitcoin has averaged about 94% annual growth over the past 5 years through to today's price of about $38,300. While being conservative? on the high-end growth rate, perhaps I am not appropriately discounting the likelihood of the high side putting it at a 20% chance. I will change that assumption in the next model. 

But just before that, let me explain why I don't think we need to worry about the mixing and matching between individual high and low estimates (e.g., stocks grow at 6.1% while bonds only grow at 1.9% or stocks and bonds are high but Bitcoin and cash are low, etc.). Assets tend to be positively correlated over longer and longer timeframes. Even though stocks and bonds enjoy some degree of poor correlation, these fade away over time as what is good for stocks (a productive, growing economy) is also good for bonds. Likewise, a world that has Bitcoin doing well probably has stocks doing well, and a world where inflation is low, stock and bond returns are also probably low. Regardless, the heart of the debate isn't going to be impacted by these details. 

Here is version 2:


Ouch! Even though I made the low-high range for stocks and bonds 50/50, the move to make low to high outcomes 95/5 for Bitcoin destroys that option. But if that is really more like the future likelihood of the outcome range for Bitcoin, perhaps stronger return possibilities on the high end are as well. So . . .

Here is version 3: 


I greatly increased the growth rate for Bitcoin using 38.6% annual growth. This isn't a randomly chosen number. This would correspond to a Bitcoin price of approximately $1,000,000, which some roughly project as a possible destination (who knows?). Regardless, stocks and bonds still look better. So let's do just two more for the sake of good order . . .

Here is version 4:


The only change here is to make the Bitcoin high possibility a little more likely moving it from 5% to 10%. And wow! Look how sensitive the difference result is to this change. Obviously this should come as no surprise as this whole thing is about Bitcoin's high end. I guess we could run just one more taking a look at a more moderate Bitcoin but also a less than total bust low-end for it.

Finally, here is version 5:


So, allowing for a Bitcoin future in any future (low end is 10% annual decline in value) gives us a fairly strong case for my gamble on some combination of Bitcoin and cash. 

Having gone through this process would I now change my mind? I will stick with 50/50 Bitcoin and cash. But that strongly suggests a question: how can I justify that choice given that I don't have an existing portfolio that looks anything like that. I am personally overwhelmingly "boring" with an almost all-stock portfolio with just a bit of crypto sprinkled in. 

At the risk of a slight digression into the post I keep promising this will not be, allow me to defend my rationality. This hypothetical is a forced gamble. My retirement investments are different in that regard. Those I can and do change periodically including both allocation as well as contribution. I get to guide those and adjust them. The hypothetical gift invested is a Ron Popeil "set it and forget it". Part of why I cannot invest more in Bitcoin (aside from it wisely not being a 401(k) option (looking right at you, Fidelity)) is that I likely cannot tolerate the variance. If I can get in and get out of it, I am as more likely to make the wrong in/out moves as the right ones. And that is before the tax-drag effect. 

Besides that, my investment reality is the retirement assets I actually do have. This hypothetical is a lottery ticket idea. If I win the lottery, my reality materially would change. I could afford more and different risk. In this sense and surprisingly, if the hypothetical was $10,000 in stocks and bonds versus Bitcoin and cash, the rational decision for me might have been stocks and bonds! Whereas the typical person would say, "that is too little to worry about the risk, let it ride!", I would counter, "I can't afford to take the the riskiness of Bitcoin at that magnitude ($10,000)." Along this one dimension, I would be right. 

I want exposure to big upsides. Unfortunately, these are difficult to find and doubly difficult to stick with. In a sense this thought experiment has revealed some of my own limitations on putting my money where my mind and heart and mouth are. 



-----


Sunday, April 24, 2022

If you've ever handled a penny, the government's got your DNA.

File this under: Wanted: new conspiracy theories—all ours came true.

When DNA testing and genomic profiling was first rolling out as a mass-market product, I remember hearing people objecting to it saying things like, "I don’t want them to have my DNA". 

These worries were summarily dismissed by science-supporting elites as paranoia on the part of anti-science or antisocial bumpkins. 

It turns out an ounce of caution here was warranted

And then COVID happened . . .


And now 23andMe has come full circle:

Wojcicki says that’s just not going to happen. “We’re not evil,” she says. “Our brand is being direct-to-consumer and affordable.” For the time being she’s focused on the long, painful process of drug development. She’d like to think she’s earned some trust, but she hasn’t come this far on faith.
Caution continues to be warranted by at least some elites (Macron refuses Russian COVID test), and I don't blame them--be sure to click through to the Atlantic story about the lengths to which the White House goes to protect the president's DNA. 

I understand Macron and the White House taking extreme precautions in this area. I also do not think it is highly likely that anything bad would come of genetic data gathering in general. In fact I tend to be supportive of the secondary (or ulterior) uses that genetic data could provide--provided there are adequate disclosures on the front end and transparency throughout the process. Trust but verify is the right approach.

The level of trust is inversely proportional to the extent to which people's fears get realized even if they are only partially realized. In other words the level of trust is directly proportional to the degree of proven trustworthiness.



Wednesday, April 20, 2022

Only Tax Unprofitable Businesses

It is income tax time again. Instead of the usual rant, I have a creative alternative for you to ponder: Only taxing unprofitable businesses. That is if we are stuck on the illogical notion of taxing income per se at all. 

Loyal readers know that we shouldn't tax corporations' profits and that we should not tax incomes in general. Corporations are just fictional middlemen between owners/workers and customers. Taxing their profits or income is just an indirect (inefficient) way to tax those owners, workers, and customers without properly changing their behavior. It would be much, much better to tax businesses as they add value to the production of final goods and services through the use of capital and labor. All of this as being part of a larger scheme to tax resource use rather than resource creation. But I digress.

Let's assume we are going to tax businesses' incomes. How should we do this? 

I propose we reverse the concept and instead of taxing a share of the revenues minus expenses (when R > E) we tax a share of the expenses minus the revenues (when E > R). In other words we only tax unprofitable businesses. 

Here is the reason for the proposal. Revenues are a measure of the benefits that a business has provided. It is an estimate of what value they have brought to the world. Expenses are a measure of the costs they have taken. It is an estimate of what value they have destroyed in an attempt to add value through their business activity. If their expenses exceed their revenues in a given year, for that year at least they have detracted from society by virtue of their activity. 

Perhaps we would need a 3-year average profitability test or carryover provision of 50% of profits from year 1 to year 2 and 25% of profits from year 1 to year 3. I'm open to ideas here. That way we smooth out business cycle and idiosyncratic effects that might otherwise undesirably punish a business in a given year for circumstances beyond its control or for investments made that have long-run payoffs. But let's not lose sight of the goal: taxing firms that cannot turn a profit (i.e., don't add value to the world).

An instant objection is that this would make a startup business unduly expensive potentially stifling economic growth. Just a little understanding of how markets work invalidates this worry. Under this arrangement a tax on an annual net loss simply adds to the cost of capital. If the expected payback is sufficient, the investment will be made. Look at it this way: Is it better that resources are used with tax encouragement (deductibility) on the prospect of future economic value creation (future profitability that would then be taxed) or that resources are used while being taxed (a discouragement to frivolous investment)? The expected return is likely the same in either case*--it is just a matter of when taxes are collected and on whom the burden falls (taxing failure or taxing success). Given that taxes discourage that which is taxed, I know in principle which one I want bearing the burden.

Thus, this method has two key attributes to admire:
  1. It punishes bad investments by taxing failure.
  2. It creates an economic environment that increases the returns to good investments by not taxing success.
A side benefit is that it potentially cleans up accounting--a lot. A great deal of effort (resources) is put into doctoring the books so that a firm looks less profitable than they actually are. This leads to an additional benefit of disincentivizing expenditure that is not actually net profitable. Executive perks, luxury offices, unnecessary equipment, etc. now all carry a burden where once they earned a subsidy. 

Does this have a chance in the hell that is our tax code of coming to fruition? Of course not. And if it did, the later temptation to reverse course would be too great to assume future profits would be tax free. But it is a fun thought experiment that yields a new way to see the economic error in our current taxing ways.







*This is kind of a reverse Ricardian Equivalence whereby known taxes on losses have to be justified by expected profits in the future. As long as the tax rate on losses is not devastatingly high, in which case it would prevent any method of transferring future expected profits to the current day to finance a current tax burden, the tax that would be applied to future profits is instead realized before those profits are themselves realized. This assumption is no different than assuming current tax rates on profits are not so high as to devastate the ability for businesses to earn a profit in our current system. 

Wednesday, April 13, 2022

Biden's Transformation Into a 1970s President

It seems clear that President Biden is well on his way to achieving his obvious goal of becoming a redux of a 1970s American president. This would be some combination of Gerald Ford and Jimmy Carter.

Consider the checklist below:
  • Stumble repeatedly ✔
  • Misspeak and garble words ✔✔✔
  • Inflation ✔✔
  • Malaise 
  • Disastrously bungled Asian/Middle Eastern war retreat ✔✔
  • Extreme tensions with Russia involving their invasion of a neighboring country 
  • High oil prices 
  • Olympic boycott ✔

Still waiting on:

Deregulation...




Saturday, April 9, 2022

Have Your House And Rent It Too

Partial list of ways my neighbors like so, so many homeowners wish to have it both ways--despite the blatant contradictions. While many of these overlap with each other, they each are distinct.
  • They want high property values, but they also want affordability--just maybe not near them? Affordability for thee but not for me.
  • They want high property values, but they don't want "outside investors" much less "speculators!" to buy properties near them--presumably to rent them to undesirables (see below).
  • They want a thriving rental market, but they don't want anyone to rent near them.
  • They want diversity and inclusion, but they champion restrictions on development and rentals which make surrounding housing unaffordable and unavailable.
  • They want diversity and inclusion, but they wish to make choices for their neighbors thwarting their personal preferences.
  • They want the freedom to make their own choices about remodeling, etc., but they do not want an anything-goes policy for even their dearest neighbors. De gustibus non est disputandum for me; no way in hell for thee!




Monday, April 4, 2022

The Economics of Immigration in One Lesson

Or perhaps I should say in one equation properly explained . . .

Political thinking: 
    ALL THE STUFF divided by ALL THE WORKERS equals STUFF PER WORKER
The implication of this is more workers means less stuff for workers.

Economic thinking: 
    ALL THE WORKERS multiplied by STUFF PER WORKER equals ALL THE STUFF
The implication of this is more workers or more productivity means more stuff for workers.

Definitions: 
All the stuff = production, the output;
all the workers = resources, the input; 
stuff per worker = productivity, the rate at which we can produce stuff

Allowing more workers (immigration as well as the removal of barriers to entry like licensure laws and minimum wages) is clearly a net good when total output rises. 

The key to consumption is production. The key to mass consumption is mass production.*
(HT: Bryan Caplan)






*Note for minimalists, environmentalists, etc.: Don't get hung up by the term "mass" here. This does not mean "excessive". It means flourishing for the masses. Limitations on production/consumption hit the worst off first and the best off almost never.