Random Thoughts on Financial Reform
By request, here are some thoughts on how I’d reform the financial system were I proclaimed High Dictator.
0. We should remain skeptical regarding efficacy of regulation to prevent crises. Financials will always find a way around regulations, creating new & unforeseen risks. W/ lots of $, they’ll always be able to hire better personnel than regulators, and thus (between that & regulatory bureaucracy) will always be one step ahead of the latter. Regulations re. size & capital are attempt to keep crises manageable.
1. Higher capital requirements: to create a larger buffer against losses. Make capital requirements counter-cyclical, escalating w/ larger size, and applicable to any firm that borrows short & lends/invests long. Also, make capital requirements the same across all types of assets; e.g., don’t have lower capital requirements for securitization bonds as opposed to loans held in portfolio.
1a. Subordinated debt: Require that a certain percentage of each financial institution’s liabilities consist of long-maturity, subordinated, unsecured debt. Debtholders thus put at risk will provide some discipline to financials, even if shareholders do not. Also, expectation of loss would make DFE swaps of such debt less disruptive, since anyone buying such debt would know that DFE conversions were a possibility. For similar reasons, we might want to consider mandating the use of reverse convertible debentures or preferred equity by financials.
3. Figure out ways to measure & limit financial interconnectedness. This would (hopefully) reduce the problem of firms of being “too connected to fail” (ala Bear Sterns or AIG).
4. Mortgage Lending: The systemic risk posed by the GSEs is obvious, as is their non-viability absent government subsidization. So liquidate them (gradually, first by slowing/halting new lending, then by either liquidating existing assets or letting them run down). Replace them with the Danish mortgage system (also here) of monoline mortgage lenders that retain & service the loans they originate, & are funded by duration-matched covered bonds. Retention of credit risk upon originator balance sheets eliminates securitization’s agency problem; covered bonds permit mortgage originator access capital markets (thus reducing costs) while shedding interest-rate & duration risk onto parties (e.g., pension funds, bond funds, long-term investors) looking for long-term investments (thus preventing another S&L Crisis). (A side benefit would be reducing the problem of negative equity.)
4a. In view of securitization’s agency problems, I’m inclined to replace it en masse w/ the Danish model of covered-bond-financed lenders; however, I don’t know enough about securitization to determine how applicable the Danish model might be to other areas besides mortgages (e.g., car loans, credit card loans). For any markets wherein securitization survives, we should require that the securitizer retain “skin in the game” (e.g., a residual) equal to the maximum probable loss on the loans in question (determined based on historical loss data from recessions).
5. Post-Lehman, the danger of runs on MMFs is now self-evident. MMFs should either accept regulation (e.g., capital requirements) in exchange for LOLR access & guarantees, or disappear. TANSTAAFL.
6. Narrow Banking: not sure how workable this is, but it’s something we might want to consider. Note that this approach would probably have to be coupled w/ licensing regulations prohibiting unregulated non-banks from entering lending markets normally serviced by banks, lest the unregulated outcompete the regulated (and subsequently implode via excessive risk-taking).
7. Prompt Corrective Action (ala FDIC receivership authority) for all financial institutions (not just banks). Include authority for mandatory debt-for-equity conversions & good-bank/bad-bank split-ups. Then use such tools next time some financial goes belly-up, so that markets realize regulators are willing to screw over bondholders rather than expend taxpayer funds bailing out financials. No longer would regulators have to rely on ad-hoc bailouts to prop up troubled financials, or rely on a bankruptcy system (e.g., Lehman) ill-suited to rapidly resolve such firms. Moreover, the existence of such authority would give notice to all potential investors & lenders, who would thereafter understand the true risks associated with lending to or investing in financials, and price such investment or lending accordingly.
8. Consumer regulations: The debacle in mortgage lending makes clear to me that consumers can’t be expected to understand any but the simplest financial products. E.g., 30-yr FRMs w/ 20% DP. So, ban all but the simple stuff that anyone can understand. The resultant credit rationing would be a feature, not a bug, with the inability to customize lending for people w/ poor credit confining credit to the creditworthy.
9. CDS: Kling-Salmon notion of treating net sales of CDS as assets on balance sheets (w/ resultant capital requirements). This would deter institutions (e.g., monolines bond insurers & AIG) from exposing themselves to massive risk by issuing tons of unhedged CDS.
10. Untried Assets & Financial Innovation: To shield taxpayers from the risk associated w/ novel financial assets, any government-insured financial would be prohibited from owning untried assets until said assets had been through a failure cycle. (Prior to such a cycle, it wouldn’t be realistically possible to determine probable losses on, and hence adequate capital ratios for, such assets.)
11. Stress Testing: As was once required, anything rated AAA should be able to survive a “Great Depression” scenario. Similar scenarios should be regularly employed to “stress test” regulated entities’ portfolios IOT determine whether they have enough capital.
12. Escalating Down-Payments: While a 10 or 20% down-payments might be mandatory for mortgages, we could require (via regulation) higher DPs for areas deemed overvalued (if, e.g., metrics like price-rent & price-income ratios are above long-term trends). IOW, cap the loan amount for a given area, such that anyone buying a house in that area must provide the difference via higher DP. By rendering homebuying less affordable in overheated areas, such a requirement would tend to reduce demand for housing in such areas, and thereby kill off housing bubbles before they started. Meanwhile, higher DP’s would also reduce the chances of homeowners ending up underwater.
A. Reducing US current-account deficit via tariffs (& threaten protectionism against anyone using monetary mercantilism on us). This includes energy independence. This would eliminate the foreign capital imports that helped finance the current crisis.
B. Higher domestic savings rate: encourage via consumption taxation. A Fair Tax is probably unworkable, but one possibility is to remove all limitations on contributions to, & withdrawals from, traditional IRAs. Keep contributions tax-deductible & withdrawals taxable. To ensure some degree of progressivity (given that the rich save more), we could cap the deductibility of contributions at (say) 50% of income for persons making over $100k.
C. Cheaper housing via deregulation of land use.
D. Kill off subsidies for housing (we already devote too much “investment” towards it); no more FHA, GSEs, MI deduction, etc.
E. Eliminate the tax-deductibility of interest payments (for both businesses and individuals). This would discourage the use of debt financing by both households & companies.