How far can finance be simplified? Part I
In the wake of the global financial crisis, there has been a lot of discussion about how finance became too complex and how it needs to be simplified. This led me to speculate on how far finance can indeed be simplified. This is a question that I would like to address in several parts as I use this blog to clarify my own thoughts. Caveat: my entire speculation is completely ahistorical – it is a clean slate design which ignores the existing institutional structure completely.
In this post, I describe an ultra-minimalist financial sector that has a payment system, a spot foreign exchange market, “pure” life insurance companies, an equity market and practically nothing else. In this ultra-minimalist model, there are no banks, no central banks, no debt markets and no derivative markets. The payment system would essentially be a retail version of a real time gross settlement system which in principle needs neither banks nor a central bank. It is essentially a piece of technological infrastructure and nothing more – a central depository for cash. Money could be fiat money or commodity money or anything else.
By pure life insurance, I mean first of all that the companies offer term insurance which unlike endowment insurance is not bundled with investment products. Secondly, level premiums would be replaced by rising (actuarially fair) premiums so that there is very little investment component in the insurance product. Finally, insurance would ideally be redesigned so that the life insurers take micro mortality risk (the cross sectional variation in mortality rates at a point in time) but not macro mortality risk (changes in life expectancy over time).
If we can make these changes, insurance companies become easy to run and easy to regulate. They simply become applications of the law of large numbers and involve vastly reduced risk taking over long horizons. Incidentally, I believe that macro-mortality risk is practically unhedgeable and uninsurable. Insurers can credibly claim to provide protection against this risk only by having recourse to a bail out by the state. Perhaps, it is ultimately beyond even the resources of the state to credibly insure against macro-mortality risk.
In the absence of debt, there are hardly any prudential regulations except possibly for the insurance companies. Financial regulation consists primarily of conduct of business regulators and consumer protection regulators.
How can a financial system operate without debt? Well, the Modigliani-Miller theorem says that lack of debt is not a serious problem for the corporate sector except that the tax advantage of debt is lost. One could assume that the government simply enacts a lower rate of corporate tax so that the elimination of tax deductible interest is revenue neutral to the state.
If there is no leverage of any kind, the need for derivative markets is vastly reduced. Corporate use of derivatives is principally to economize on equity capital. Since equity is the ultimate hedge of every conceivable (and inconceivable) kind of risk, if you have enough of equity, you can choose not to hedge anything at all and still you will not go broke. One could use a version of the Modigliani-Miller argument to show that corporate hedging is irrelevant unless it introduces deadweight losses like bankruptcy costs.
This is not the whole story because apart from corporate hedging we must also worry about optimal allocation of risk among individuals. I believe however that in the spirit of Arrow’s 1964 paper (“The role of securities in the optimal allocation of risk-bearing”), the equity markets span sufficient states of the world to permit a reasonable allocation of risk bearing among individuals. The practical consequences of not having a derivative market may not be much in a world without debt.
General insurance is essentially a substitute for derivatives and it too can be eliminated in this minimalist design. If corporations do not have debt and if individuals hold diversified portfolios of non human assets, then they can self-insure all forms of property risk. Insurance is required only for non diversifiable human capital which is why pure life insurance cannot be eliminated.
The Capital Asset Pricing Model would not hold because there is no risk free asset. I do not see this as a problem because we would still have the zero beta model of Black (1972) which is not significantly more difficult to use. (As an aside, if we focus on real returns instead of nominal returns, there is no risk free asset anyway. Inflation indexed bonds issued by the government are subject to significant default risk since the printing press is not sufficient to pay off these bonds.)
Passively managed mutual funds (exchange traded index funds for example) are nice to have because they allow individuals to achieve diversified portfolios very easily. But if the stock exchange allows shares to be traded in small lots, even small investors may be able to hold 25-40 stocks directly and obtain most of the gains of diversification. Exchanges could also allow basket trades in indices to achieve the same thing.
In the ultra-minimalist design there is no trade finance other than whatever trade credit one corporation chooses to extend to another out of its own resources. There are no letters of credit (which are actually highly complex credit derivatives!).
The absence of a debt market means that there are no mortgages. One possibility is that most houses are owned by corporations that rent it out to individuals. We do not need to own the homes that we live in any more than we need to own the offices that we work in. (See my post on this a year go.)Individuals would be able to obtain exposure to real estate by buying shares of these companies. (Another – less preferred – option is that people would live in rented houses in early stages of the life cycle and buy houses only later in life.)
One difficulty with the minimalist design is the lack of educational loans. Education would have to be financed through equity claims to an even bigger extent than it is today. Even today, the successful university that attracts lots of endowments essentially has an equity claim (an out of the money call option?) on the human capital of each of its alumni and the returns on this equity claim are used to subsidize (finance) a lot of the education.
I see two big problems with this ultra-minimalist design. First is that Jensen-Meckling pointed out in 1976 that the disciplining power of debt is useful in reducing the agency problems between the managers and the shareholders. We would therefore need very robust corporate governance frameworks (including a healthy market for corporate control) in a world without any debt. This would be a problem more for mature businesses that generate a lot of free cash flow.
Second is that under the assumption that governments are and will always be profligate, we need a government bond market. Theoretically also government debt can achieve some intergenerational transfers that would otherwise be difficult to accomplish, but I think this is less important than the practical reality of governmental profligacy.
In subsequent blog posts, I hope to extend the ultra-minimalist financial sector to allow for a corporate and government debt market increasing complexity one step at a time.
On the whole however, even the ultra-minimalist financial sector would be able to support a vibrant and modern economy. Clearly, it is the equity market that does the heavy lifting in this minimalist design by taking charge of both resource allocation and risk allocation.
Posted at 6:14 pm IST on Tue, 7 Jul 2009 permanent link
Categories: post crisis finance
Comments