Since the financial crisis of 2008, everybody and their mother has been looking for some way to make sure it doesn't happen again. The responses so far have been woefully inadequate. No one thinks the reforms that have been enacted or are being considered would solve the problem.
The most "radical" reform anyone has proposed is a return of the separation between retail and investment banking, which in my view doesn't solve anything: the two biggest busts in the financial crisis, Lehman Brothers in the US and Northern Rock in the UK, were full-investment banks and full-retail banks respectively. What's more, the devil is in the details: when you get down to it, it's very hard to draw the line between "retail" services and "investment" ones, or at least not in a way that would choke credit off the economy and take the financial system back to the 1990s. So a "New Glass-Steagall" would be at best an irrelevancy, and at worst a nightmare.
So, what do we do? I've been thinking about this on and off for about four years, and I finally have an answer that's pretty close to satisfying me. I'd like to submit it to you in the hopes of refining it (or abandoning it).
My blueprint has two basic planks:
- A return to the partnership model
- Almost complete deregulation of the financial system
I know, I know, but hear me out.
What should be the goal of financial reform? Its goal should be not to prevent bubbles and busts, which are the normal result of an economy full of "animal spirits" (quiet, the Austrians in the back!), but to prevent the busts from a) necessitating taxpayer bailouts and b) having ripple effects that threaten the very existence of the financial system and wreck the economy, and by the way c) still ensure that credit flows throughout the economy (i.e., don't destroy the village in order to save it).
The question we need to ask is not why there was a real estate bubble, but why this bubble threatened to annihilate the world as we know it. My basic mental model of the financial crisis is that it became this devastating thing for basically two reasons ran amok: agency and scale. Agency (the principal-agent problem) meant that everyone knew the system was crazy but no one did anything about it. Subprime mortgage brokers knew they were selling mortgages people couldn't afford. Investment bankers knew they were packaging up products that would blow up. But it was always going to be somebody else's problem. This is what set us up for the collapse. And scale meant that the collapse of one large bank threatened to bring down the entire system.
Again, we don't want a perfect system, we want a resilient system, i.e. one which will be able to handle shocks without crumbling. Under a resilient system, financial crises would not look like October 2008, but like September 1998, or the collapse of Long Term Capital Management. The fate of LTCM is well-known to most finance wonks, but not by the broader public, and that's the point. What was different about LTCM? LTCM was big and interconnected enough that it needed to be bailed out; but it was small enough that it could be bailed out by the private sector (i.e., creditors) and not taxpayers. The failure of LTCM jolted the financial markets and ruined many careers, as it should have, but didn't cause an economic collapse, or even a downturn.
(This is why I care about scale more than interconnectedness--it's interconnectedness that means some institutions have to be bailed out, but it's scale that makes the bailouts dangerous politically and economically.)
So, how do we make sure the next institution to go down is an LTCM, and not a Lehman?
As I said above, we need a structure that solves both the agency and scale problems. And the partnership model seems to be it. The partnership model holds that stockholders in a financial institution must also be employees of that institution, and that they have unlimited liability in a bankruptcy.
It's easy to see how it solves the agency problem: the people providing the equity capital, the people running the ship and the people who are on the hook when the ship sinks are the same people. Will this prevent mistakes from being made? No, humans are foolish and greedy sorcerer's apprentices, and we'll always figure out ways to make stuff blow up in our face. But it gives everyone the incentive to be more careful. Bubbles would still happen, but people would drink less kool-aid. It doesn't guarantee they won't happen, but it's the best incentive structure to limit their destructive effects.
And it also solves the scale problem: without access to the public or even most of the private markets (I should like partnerships to be able to have up to 20/25% of their capital open to outsiders) to expand their equity capital, banks would necessarily be much smaller. Goldman Sachs' balance sheet today has about a trillion dollars' worth of assets in it. That's simply impossible under the partnership model with equity capital provided by employees--indeed, it's the reason why Goldman Sachs, which was under the partnership model until the late 1990s, went public, at the cost of its client-focused culture. Instead of a $1 trillion frankenmonster, you would have ten $100 billion mini-Goldman Sachses. If one went bust, it would be much less of a deal.
Ok, so if it's that easy (setting aside for now the political problems that would be involved with a law that would cause every large bank to break up), how come people haven't already thought of it?
Let's work through the main objections to the partnership model:
It's antiquated. You say potato, I say potato. You say antiquated, I say resilient. The partnership model dominated the world of finance up until the 20th century. It financed the Age of Exploration and the Industrial Revolution. That ain't too shabby. It's like your Grandma's old clunker: it doesn't have air conditioning or power steering, but by God, it'll get you from point A to point B whether rain, sleet or snow.
What's more, it's worth noting that Swiss banking is still dominated by the partnership model. Swiss private banks boast on their website that your money's safe with them because their own money is on the line with yours. Not all of them: there's UBS which tried to go the Goldman route and blew itself up in the process. The partnerships, meanwhile, did much better through the crisis. There are many words people use to describe Swiss finance, but "antiquated" is never one of them.
You need megabanks to provide some sophisticated financial services. When people say the banks should shrink, this is the counterargument. And it's always raised by the megabanks themselves, which leads me to discount it. But let's think about it.
Let's take the textbook example of a "socially useful" use of financial derivatives: helping an airline hedge its fuel costs. It's a complex operation, and if you want to do it with one bank, that bank should probably have a large balance sheet, offices in several countries, and so on. Ok. But making it cheaper for American Airlines to hedge its fuel costs didn't save it from bankruptcy. I'm not being flip, because it would probably make some services harder to get and we should care about that, but would it really be so bad?
What would happen in practice is that AA would have to get its hedge from several banks instead of one. That would make it more costly, for sure. But it doesn't seem like it would make it much more costly, or at least certainly not so costly that it's worth the price of a Lehman-style event every 10 years. What's more, AA hedging with several banks would not just be costly: it would also be clever, because it would increase competition and decrease counterparty risk. After all, hedging with one bank may be cheaper, but it creates huge counterparty risk. You may not care about that if you think that your counterparty will get bailed out if it blows itself up, but that's the point: the logic of the system perpetuates too-big-to-fail.
Let's take another service that only megabanks can supposedly provide: prime brokerage and market making. If you want hedge funds and other investors to be able to access every instrument under the sun, get leverage and so on, the logic goes, a prime broker and market maker needs to have a huge balance sheet to be able to be a counterparty to those transactions. It makes intuitive sense, but I'm not so sure. After all, Citadel Group, a hedge fund, recently started a prime brokerage and market making business. Citadel is one of the biggest hedge funds in the world, but it's a minnow compared to the megabanks--again, more LTCM than Lehman. Even Bloomberg has quietly built a big brokerage business (and it's worth noting that Bloomberg is incorporated as a partnership)--Bloomberg is an agency broker, not a market maker, but it goes to show you may not have to be a megabank to be an intermediary in the market. And as with AA above, hedge funds have more than one prime broker anyway (at least since 2008), so they should be fine.
Being a prime broker and market maker with a huge balance sheet is useful for one thing: having an unparalleled view of the market and being able to trade with an information edge over other people (including your clients). But the preservation of Goldman Sachs' most profitable business should not be the goal of financial reform. (Nor should it be "punishing" the big banks for the sake of punishing them!)
Finally, there's a third objection to the partnership model: that it would choke off the supply of credit.
And, well, it's hard to think that it wouldn't, at least in the current financial system. Banks lend against capital, and the partnership model would drastically reduce the amount of capital they have available to lend against.
It would also make bankers more risk-averse. When people talk about financial reform, they often talk of the Good Old Days when you and your banker had a relationship and the banker would only lend money to people he (and it was a he) personally knew. How good those days were. They were also days when access to credit was dramatically curtailed. What you might call the industrialization of finance, like industry itself, has created industrial accidents and pollution, but it has also made its product--credit--cheaper and more widely available than ever before, which has been, still, overall, a blessing to humanity.
In this light, it's hard to see how a return of the partnership model, just by itself, wouldn't squeeze credit massively. If you have more to lose, you will lend less, by definition. The Good Old Days of banking were Good Old Boys days; if you were in the club, you could get funded, and if you didn't, you had to do without credit.
And as I said above, one of the goals of financial reform should be to make credit widely available. What then?
This is where my second plank comes in: massive deregulation of the financial system.
It sounds scary when you put it like that.
But let's think about what regulation is supposed to do. The main objectives of financial regulation since the 1930s have been twofold: a) to give retail investors (a.k.a. "regular folks") trust in the system so that they would feel comfortable putting their savings in the market, and b) to prevent catastrophic collapses. Since the 1930s, the government has said to banks: ok, we'll insure your deposits (i.e., bail you out), but in exchange we tell you how to run your business so that you don't collapse so that we don't have to bail you out.
I think a) is misguided. I think it was The Epicurean Dealmaker
who said on Twitter that the mandatory disclosure/warnings for anyone who signs a financial document should be a slap in the face and the phrase "CAVEAT EMPTOR!" shouted into their face drill instructor-style. We don't want retail investors to have trust into the system. Finance is like fire: it can be very useful, but it burns and can be very destructive. The last thing we want to do is give people the impression that we can somehow regulate away the bad consequences of fire. And of course, despite all the regulations, on Wall Street the deck really is stacked against the little guy. Your sell-side advisor and mutual fund's job is to nickel-and-dime you with fees so that you end up with lower returns than if you'd bought the market.
As for b), preventing collapses, remember--thanks to the magic of the partnership model, banks are no longer systemic. We could mandate that people buy deposit insurance to avoid bank runs, but that's it.
Ok, but why is it so important to deregulate? What does this have to do with the supply of credit?
Everything. What causes markets to be underserved is quite often a lack of competition. The fact of the matter is that banks are highly regulated, lending is highly regulated, and so the sector is not competitive at all. The barriers to entry are very high, and so banking is basically an oligopoly.
Here's the deal: I should be able to start a bank. My expertise is at the intersection of finance and online business. I should be able to raise some money from family and friends and start a bank that caters to online businesses. I might do "retail" banking like taking deposits and lending, but I could also do investment banking: corporate finance advisory, securities underwriting, equity investment and so on. I could provide wealth management services to the people who run online businesses and often become very rich.
Why can't I do that? The answer is that the phonebook looks like a brochure compared to the number of regulations I'd have to comply with before I could even start. If you look at the relatively unregulated aspects of finance, like hedge funds and private equity, they're fairly entrepreneurial: people are always striking out on their own to start funds. No one's striking out on their own to start a bank. There's no reason. Everyone seems to think that a bank should be a staid, buttoned-down place founded in the last 19th century, but that's only because it's always been that way. (My reform would limit the liability of bank partners for the first 5 years of operation precisely to encourage people to strike out on their own.)
(My hypothetical startup bank example is also why a "New Glass-Steagall" doesn't make sense to me: my startup bank would obliterate the wall between retail and investment banking, and so what?)
This isn't just an academic problem: every startup that has tried to improve finance has run into regulatory roadblocks. PayPal was almost shut down by state regulators in the early 00s. Simple
, previously known as BankSimple, languished in limbo for a long while until it could figure out a way to provide its services in a way that would satisfy regulators. Lending Club was shut down by the SEC for months. Square has large compliance staff.
It's really hugely important. In some sectors, like, say, trucking in the 1980s, deregulation merely makes things more efficient. Some incumbents lose, some new entrants win, things are cheaper, but overall the market basically remains the same. In other sectors, deregulation has the potential to transform everything.
I think this is true of finance. As I wrote in a research note
, the way companies are getting financed is completely changing. And the companies springing up to serve those needs are being stifled by regulation whose logic is still stuck in the 1930s.
In the UK, Wonga
uses very good algorithms to provide short-term credit to tons of consumers who couldn't get it before. It lends from its equity capital, not deposits.
In the US Lending Club
aggregates borrowers and lenders in a marketplace. When you want to borrow money, it uses clever algorithms to estimate your risk and gives you the appropriate interest rate. On the other end, people can buy those income streams at the risk/reward ratio that they like. Lending Club uses the tools of "casino finance"--it's a marketplace, and it essentially securitizes loans--to expand access to credit and investment opportunities. It does what banks should do, turn savings into credit, but does it very efficiently.
In the United States, it is onerous for all but the largest firms to list securities on the public markets. Why is that? Well, regulatory compliance costs are very high. It's not the only reason though. You also need to get investment banks to underwrite your offerings, and small companies are too small to make it worth their while. But the reason why there are very few investment bank boutiques now is because banking has become an oligopoly, and that's because of regulation. Going even further: why should you need a bank to underwrite your offering? Well, because they have the connections to the big funds that are going to buy your offering. But in a truly liquid and open market you don't need that. Do you need to phone up some sort of intermediary before you can sell your lawnmower on eBay?
In Europe, there have been encouraging results of stock exchanges that have allowed small and medium-sized businesses to list debt obligations, and of funds that securitize small and medium business debt to sell it to funds.
Whatever the mechanics, the point is that the benefits of accessible and liquid securities markets can and should be available to all businesses, and that this would be a tremendous boon to society. And information technology allows us to do that. What doesn't is regulation. With a service like InDinero a business can have its financial history in the cloud. It's not hard to imagine that a small business owner could go to a Lending Club-like marketplace, let it access its data via a Facebook Connect-like service, and be able to sell debt into a marketplace. This would be a boon to the world.
Meanwhile, bank-owned regulators prevented Walmart from offering financial services, even though the main beneficiaries would have been low income Americans. In a deregulated system, many other forms of banking could emerge: companies would offer banking services (after all, car companies are basically banks), credit unions and other initiatives would blossom and so on.
So, what does this all have to do with partnership banking and the supply of credit? Well, everything.
Even though the US is overregulated and access to credit is insufficient, it is still much better for businesses than Europe. Why is that? Because a much bigger amount of credit in the US comes from financial markets, not banks. Therefore US banks, despite being an oligopoly, do have to compete more with financial markets to provide credit to companies, and so they have to align.
Force the partnership rule on the current financial system, and credit would shrivel. But force partnership banking on a system where non-bank and new forms of credit are widely available thanks to deregulation and where entrepreneurial banks can get off the ground easily to serve market niches and opportunities, and it seems at least possible to me that credit would still flow, and be allocated much more efficiently, with much lower systemic risk.
So, to summarize:
- Let a thousand flowers bloom, let people do all sorts of clever financial wizardry, encourage people to become financial entrepreneurs;
- https://www.wonga.com/ Mandate the partnership model for banks to prevent system-damaging explosions.
You can follow me on Twitter here.