Creating a Bull Market for Government
Why a Terrible Idea for Wall Street Might be a Great Idea for Government
I have a deep and long-term interest in both distance running and collateralized debt obligations. I read far too much on both subjects. Running is about the benefits of simplicity and CDOs are about the dangers of complexity. I suppose I am drawn to these topics because of the intrinsic purity that lives at the extremes, the purely simple and the purely complex. This also explains the popularity of Malcolm Gladwell and books like Outliers and David and Goliath, which are essentially about rare positive events, and Nassim Taleb’s Black Swan which is essentially about rare disasters. My tastes are hardly unique, as they reflect a pop culture that absorbs serial killer mysteries and Touched by an Angel episodes in equal proportions.
But I am also interested in what is right next to those pure things. For instance, right next to running books are endurance books, and at the top of this list is Alex Hutchinson’s Endure: Mind, Body, and the Curiously Elastic Limits of Human Performance. The central thesis of this 2018 best-seller is derived from the thinking of Dr. Tim Noakes, a widely respected exercise scientist, who is the progenitor of the Central Governor Theory. That theory posits that the mind puts brakes on your own performance by sending overwhelming signals when you have reached your limit that you should stop immediately. Endure proposes that those signals from your brain can be ignored because your mind is far too conservative in its beliefs about your body and your ability. Hutchinson argues that the greatest endurance athletes routinely ignore their brain’s signals to stop. The metaphysical and epistemological implications of this theory or mind-bending. Because, wow. And because if your mind doesn’t accurately know your own risks how can you use your mind to overcome them?
Anyway and obviously, while holding your breath for many minutes or running over 3,000 miles in a race around a single city block in NYC is possible, it is not a terrific idea.
By the way, as Jonah Hill says to Brad Pitt at the end of Moneyball, this is a metaphor.
Anyway, as you can see, this adjacent literature is interesting and provocative. Let me give you a provocative idea of my own from the edge of the other literature I read obsessively, the collateralized debt obligation literature.
I’m going to write about bookies and Wall Street and debt collection and leg-breaking and then I’m going to tell you why this is an interesting business for government to get into and why it could be a huge boon for people who care about helping other people. Just keep in my mind that this isn’t a real proposal, just an idea of a different way of thinking about financing services for hard-to-serve populations. The goal is to highlight that we desperately need better ways to finance services for prevention and that maybe some creative thinking can move the dial a bit.
What Your Bookie and Wall Street Got Right
A CDO isn’t that complicated, it’s just a way to smash together a bunch of debt and resell it as its own thing. ‘Securitization’ is just the process of taking that debt and turning it into something that can be bought and sold on its own. But the terms are unfamiliar to most people and off-putting, so let’s think about it instead from the perspective of the bookie who lends money to gamblers down on the docks.
If you are a bookie, you might look at your portfolio of stevedore gambling debts and think, man I spend way too much time down on the docks trying to get my money back from these guys who just jump on a ship and head for Gdansk. I think I will take a bunch of these bad debts and put them all together and sell them for pennies on the dollar to another bookie for collection. Your reason for doing this is obvious. The other bookie might enjoy breaking legs more than you do and he might see the package as cheap. So those are debt obligations you have ‘securitized’—you have packaged them together and sold them off.
But now imagine a world where there are lots of bookies and they all know each other. And now they start looking at each other’s portfolio of debts and maybe start thinking hey, John just bought a bunch of debts and I know some of the guys who owe that money. I think they will pay so maybe I should buy this debt at 50 cents on the dollar from John. Then, some guy in LA says, hey, I know John by reputation, he runs a bad bookie shop and is terrible at collecting so he is probably undervaluing these debts. I want to buy those debts and I will pay 60 cents on the dollar, even though I know nothing about the actual debts, I just know a little about how John runs his business. Now there is a market for the exchange of these CDOs.
Now imagine that the bookie in L.A. is really smart and says, hey, I think I’m making a great deal here but it would be an even better deal if I bought some insurance. So while LA bookie holds onto the debt, he sells the risk of default to another bookie in Dallas. UIn that exchange, the Dallas bookie agrees to pay the LA bookie if all the debts go bad (which neither think is likely) and the LA bookie agrees to pay the Dallas bookie a monthly premium until all the debts are paid off. What I just described is actually a credit default swap, and this is actually what I want to talk about.
Why Credit Default Swaps are not Inherently Evil
Credit default swaps are both super simple and a little tricky. They are simple because they are just another type of insurance. They get tricky because anyone can be the insurer and anyone can be the insured. To understand it, you have to put aside whatever preconceived notions you have about insurance and really think about the essence of the transaction. Don’t worry about what label you put on the people involved, just think about what they do in the deal.
When you buy car insurance, you are essentially going to Allstate or Nationwide or whoever and saying, hey, I’m probably going to wreck my car, will you buy me a new car when I do? And Allstate looks at their actuarial table and says, we don’t think you will wreck your car so sure, we are happy to sell you insurance. And so you make a deal where you agree to pay them a steady stream of dollars. And if you don’t wreck your car, and you likely will not, they just put that money in their pocket.
And it might go even further where you say, hey, Allstate, I think I’m going to die soon sell me some life insurance to protect my family if I die. And Allstate looks at their actuarial table and says, we don’t think you are going to die any time soon, so sure, we are happy to sell you some insurance. And you pay the premiums which they put into their pocket.
Incidentally, both Warren Buffet and AARP are mainly in the insurance business along with a bunch of other people you probably don’t think of as insurance people. It’s a lovely business because the cash comes in today and the expenses are down the road. That’s part of why I think this model is appealing to both government and philanthropy. But read on…
The key is that the essence of the transaction is that you think something bad is going to happen to you and your insurance company thinks everything is going to be ok. Insurance then is where you bet against yourself.
Why Swaps Make a Market Explode
What the credit default swap does is generate huge amounts of liquidity around these transactions. Every transaction involves some risk, and reducing that risk effectively lowers the price and creates much more demand for whatever it is that is now less risky. On the other side, purchasing these swaps creates huge cash flows and those dollars can be put to use. If you are trying to grow a stagnant market, like say, the market for prevention, that is an appealing recipe.
And if you think about it in those terms, it’s really no different than when you buy car insurance or life insurance. As odd as it sounds, if you couldn’t bet that you were going to wreck a car, you would be less likely to buy the car in the first place. Without people betting against themselves, the car business and the home buying business would grind to a halt.
This is essentially what happened in the mortgage market. Before credit default swaps, lenders who sold mortgages could not bet against their own portfolio—they couldn’t insure against the risk that the homeowners they lent big bucks to would not be able to pay. Before credit default swaps, lenders had to hold on to the risk within their own portfolios. And as a result, they were very, very reluctant to lend to anyone who did not have exceptional credit. And they made up all sorts of terrible rules about who could even have exceptional credit with disastrous consequences. Credit default swaps changed all of that and made the mortgage market much more liquid and opened up homeownership opportunities.
Now let me stop there. All of that should sound ok to you. There’s a lot of moving parts and fancy jargon in the CDO and credit default swap world, but at the level I am currently talking about, everybody is clear on what they are buying and selling. Everyone has a chance to fairly evaluate their own risks and rewards. Now, I’ll say a little at the end about what went wrong in the mortgage market because the world I just described is really Wall Street in the mid-1990s. By 2008, the market had gotten vastly more complicated and divorced from reality and that’s when the music stopped. But, if regulators had stepped in right here, right at this point, this market structure would have been imperfect, but ok.
Liquidity and the Mortgage Market
All of this risk buying and selling creates liquidity in a market, which is generally a big positive. But it is important to note that liquidity is not a noun, it is just an idea. The idea of liquidity is that there is money around and available to be moved from wherever it is to wherever you want it to go. That’s it. If you have tons of money in your retirement account and your kid is heading off to college, you might have a liquidity problem. You have the money, but you can’t get at it without a big penalty. So, you’ve got to find some liquidity by moving things around or borrowing money or your kid is not going to college no matter how fat your IRA.
A big barrier to homeownership in the US has historically been a lack of liquidity among lenders. When your mortgage was just a one-time, one-off deal between you and your bank, the market was pretty inefficient and there was a lot of friction. If you were a person of color, America intentionally created an outrageous amount of friction—so much that Douglas Massey and Nancy Denton said the friction created an American Apartheid. Policy efforts—ending redlining and creating new incentives—made a dent in the illiquidity of non-white homeownership, but not nearly enough.
Many, many progressives saw these jargon-y financial instruments—particularly credit default swaps (but note that credit default swaps don’t work with the CDOs)—as a way to create a tremendous demand for mortgages to be issued to people that had been traditionally shut out of the financial markets. And they did! The subprime housing market exploded and made massive numbers of loans to people who never would have qualified for them in the past.
Creating a Market for Prevention
So, what can we learn from all of this that is helpful to governments looking to serve hard-to-reach populations: the disenfranchised and the disadvantaged? The ability to buy insurance creates demand and liquidity, and this allows markets to grow much faster than they otherwise would.
What government does not have today is a way to hedge its bets. And so it doesn’t bet at all. It just does the same thing endlessly. The government needs to buy insurance. It needs the ability to swap its risk of making bad bets.
The thing about credit default swaps that makes them such a massive improvement over insurance from the government’s perspective is that credit default swaps are endlessly flexible. The arrangement doesn’t have to be one where the government pays an insurance company to take on its risk. It could be the opposite. It could be backward. It could be one where a socially-minded impact investor or big national philanthropy buys the insurance rather than selling it to the government.
Suppose a consortium of national philanthropies thought it was extremely important to develop very specific skills—STEM skills—to the millions of children in low-performing schools who receive too little training. Suppose that these philanthropies were convinced millions of lives could be improved by doing so. And suppose these national philanthropies were convinced that they can not do it on their own. Suppose these large philanthropies know that their endowment is too small to meet this challenge. (None of this is hard to imagine, by the way).
If the idea is to pump $1 trillion into STEM education to help American students be competitive on the world stage, regardless of whether they are poor or rich, native or immigrant, it’s going to need the government front and center. The benefits to the government are innumerable: greater tax revenue, far fewer problems to remediate, a jump start to countless young lives. So, the challenge to philanthropy is how to get the federal government to take on this project. But the government hasn’t done this yet and almost certainly won’t do this without a huge push. And a big problem is that the government is relatively illiquid—it’s like you when your kid is headed off to college—it has tremendous assets but they are all dedicated. And the idea of borrowing to fill this gap is unrealistic in today’s (or yesterday’s or tomorrow’s) political environment.
That’s the challenge for philanthropy. How can philanthropy invest in a way that gets the government to invest alongside it at the scale that is needed?
The idea here is for philanthropy to turn its usual approach to investing on its head. Usually, philanthropy tries to get the government interested in making an investment by first seeding the field and demonstrating that an approach has merit. And, that the approach is low risk.
What if instead of seeding the field and then enticing government to follow philanthropy into the field, what if philanthropy decided to help the government make the investment itself? What if philanthropy directed its resources toward the government?
Now, everything discussed above comes into play. What the government really needs is insurance so that it can hedge its bets and create a more liquid marketplace. The market for prevention is static and illiquid, demand is limited to a government monopsony (for the most part). Insurance, as we have seen in the mortgage market, can vastly improve the efficiency of a market. Perhaps it can do the same for prevention.
Here is where the credit default swaps serve everyone’s interests. The idea is that philanthropy wants to buy a ton of STEM in the same way you want to buy a car or a house. If you want to buy a car, you need insurance to reduce your risk so the investment is worthwhile. The government needs some kind of insurance in order to buy into the prevention market. And it needs the liquidity that comes along with a lower risk marketplace.
But here, it is not really the government’s interest that matters, it is really philanthropy that has the interest. Philanthropy wants the government to buy STEM. One way to do that is for philanthropy to offer insurance to the government. But that doesn’t work, because if the whole thing fails philanthropy cannot possibly come up with the trillion dollars to make the government whole. Plus, the philanthropy business model is to make investments. Enter the credit swaps.
I want my 16-year-old son to buy a car so he can do some of the driving to help out the household. But he can’t afford it. I don’t want to just buy the car and let him drive it. Buying the car does not solve my household’s problem, which is that we need a third driver. Owning the third car adds to the household problem.
So I make the following proposition to my son. I will pay the down payment on the car if he picks up the payment. But he can’t really afford that either. What he proposes in return is that he works off some of the value of the car by doing errands for the household and he pays the insurance. But that leaves me basically where I started, which is owning a third car I don’t want.
Now imagine that insurance is not required in our state (which is certainly true for investments in STEM, there is no insurance to buy). So I offer him a different deal. I tell him that I will sell him insurance for the car. He will pay me a premium, and if he has no accidents, I will apply those payments as the car payments and he can slowly buy the car from the car company, with me as an intermediary.
Now everyone is happy. I have a 3rd driver and some relief on running household errands, but I don’t really own a third car. My son has a car he can afford and there are big disincentives for him not to wreck the car (normally if he wrecks it the insurance company gives him a hard time, in this scenario his Dad kills him). America is happy because a car was sold that otherwise would not have been.
This deal between me and my son is essentially a credit default swap.
It is also a plausible model for how philanthropy gets the government to invest big in STEM and other prevention. The way this can work is if the government agrees to provide the start-up capital for the STEM investment. The great thing about STEM investments is that they pay for themselves over time. Tax revenues go up, and costs go down as needs decline. But all of that is a little risky because there is only a little bit of a precedent for this kind of government investment (the social impact bond).
The solution is for philanthropy to step in and sell the government credit default saps. Typically the seller of the swaps is the one who owns the risk, and that is the case here. In selling the swaps, philanthropy agrees to pay the government a premium to take on the risk of the whole venture. In this case, the risk is not that the whole thing will default, but rather that the stream of benefits will be smaller than expected. The premiums paid by philanthropy are used to cover any shortfalls by the government in the expected return on the government’s investment.
In practice, the government would set up a quasi-public financial authority, similar to a stadium authority or a transportation authority, which it calls the Prevention Authority. The government seeds the PRevention Authority with XX dollars and those dollars remain the government’s funds so there is no risk. Philanthropy, through the Prevention Authority, designs the programs and implements them. The government’s role is simply as an intermediary, as the link between the people who really need these services and the people who can get the services to them.
The Prevention Authority invests the government capital in a prevention structure and uses the stream of revenues (the insurance premiums) flowing into it to create liquidity in the market as needed, and of course, to make the government whole if the benefits fail to materialize. All the government needs to do is assume liability if it doesn’t work and that will be very low risk if the Prevention Authority focuses on evidence-based best practice. And in fact, the government will assume far less risk than it does today.
This financing mechanism also creates tremendous transparency, the transparency that does not exist in the present system, The government and the Prevention Authority would have to loudly proclaim any failure as part of the deal. The Prevention Authority either dramatically improves STEM outcomes among poor children in bad schools or it doesn’t. And it has to tell us which it did.
The number one area where there is too little liquidity today, where there is too little investment, is prevention. I have a little piece here that describes why the government invests so little in prevention. The need for greater financing of prevention—from reducing child poverty, to improving socio-emotional learning in schools to preventing substance abuse relapse to STEM—is massive. Credit default swaps are a mechanism to dramatically improve the level of investment.
With enough liquidity through this risk-sharing, the Prevention Authority could create a very big market for prevention.
Anyway, I should close by explaining why the failure of the mortgage market and the collapse of the US economy as a result of the widespread use of credit default swaps and collateralized debt obligations is not a red flag for this idea. Because it is not.
As I have noted, the mortgage market circa 1996 made a lot of sense and was fairly stable and serving important policy goals. The central problem, however, was that the market was not developed with altruistic objectives but with greed as the motive. Greed begat greed, and the derivatives metastasized and overwhelmed the financial system. Derivative ideas like synthetic CDOs were spun up and out of control, ideas which required the creation of bad debt—literally required that mortgages were given to people who could not pay them back. Greed and madness.
From these ashes, there is a great idea, if taken completely out of that context and applied with better, perhaps even pure, motives. It’s just one idea, but the broader idea, that we need to think big and creatively about how we finance prevention and finance policies and programs for hard-to-serve populations, certainly needs a gigantic jump start.
So this my contribution to that process. A Tesla rocket that shoots off into the sky, gathers tons of critical data, then rushes too quickly back to earth and explodes on impact. But with enough data to declare it a success.
A Big Story, in Two Charts
Andrew Gelman is amazing, and you should read him regularly. This morning (December 20) he wrote about the grand debate that is now emerging about the Heckman Curve. The curve essentially argues that investments in early childhood and adolescence are far more beneficial than investments later in people who are later in life.
But, he notes a paper from David Rea and Tony Burton, New Evidence on the Heckman Curve, that makes a strong counterargument, in one delightfully uncorrelated scatterplot.
Rea and Burton note, “[o]verall the extent to which a social policy investment gives a good rate of return depends on the assumed discount rate, the cost of the intervention, the interventions ability to impact on outcomes, the time profile of impacts over the life course, and the value of the impacts.”
That makes it quite obvious, for instance, that a low-cost intervention with modest benefits for a child that persist throughout their life would seem to ‘dominate’ in policy terms a much more costly intervention in adults, even if the adult intervention returns a substantial benefit. My take is that we are learning that the intervention with children is more likely to focus benefits within the child whereas the adult intervention may have spillovers across their community if, for instance, it reduces violence.
There’s also a healthy skepticism off all parties in this debate about the value of meta-analysis, which tends to gloss over threats to validity within individual studies.
In sum, it is probably worthwhile to set aside the Heckman Curve as the default in decision-making heuristic and to take a more nuanced approach.
Musical Interlude
So, prevention, I have found, excites very few people. It perhaps conjures images of getting a shot at the doctor’s office or receiving a stilted lesson in school. It is not that. It is about human dignity, about helping someone to find a meaning and a purpose. From Guatama Buddha, “If you do not change direction, you may end up where you are heading.” Or here in this rather extraordinary video, and more succinctly, you are either headed somewhere or ending up somewhere.
Quote of the week
“O'Hare is the worst. It smells like stale Olde Style and sadness” - Kevin Krause.