Regulation T regulates the extension of credit by broker/dealers including the percentage brokerage clients can borrow against their securities in margin accounts. That initial margin is currently set at 50%.
In the late 1990s, when then-Fed Chairman Alan Greenspan spoke about the "irrational exuberance" of stock (and other asset) prices, I wondered why he didn't recommend tightening the margin requirements governed by Reg T in a counter-cyclical fashion. Perhaps the reason he didn't advocate this was because he was wary of trying to determine when asset prices are high due to "irrational exuberance" and when they may be high because fundamentals warrant it. A way around this might be to use average valuation metrics over long periods of time as a guide: for example, market average P/E ratios over the last century, or, to account for earnings cycles, market average P/E 10 ratios (price divided by the average earnings over the previous ten years). Using such metrics, margin requirements could be adjusted on an annual basis, e.g., set at the current 50% when trailing P/E ratios are 15 (about the hundred year average), set at, say, 60% when trailing P/Es rise to 17, and set at, say, 40% when trailing P/Es drop to 13, etc.
That's one idea of a counter-cyclical policy to moderate extreme volatility in asset prices. Another might be to invest part of the Social Security Trust Fund (currently invested in a special class of Treasury securities) in stocks, corporate bonds, or other non-Treasury assets, guided by similar valuation metrics (e.g, new Social Security Trust Fund monies could be invested in stocks only when trailing market average P/Es were below 15). The idea of investing part of the Social Security Trust Fund came up in the late 1990s as well (when market average P/Es were of course much higher, since this was the tail end of the secular bull market in stocks that began in 1982) when it was mooted by the Clinton Administration. At the time, Chairman Greenspan opposed the idea because he worried about the possible negative effects of government intervention in the capital markets, and the idea was never implemented.
I suspected at the time that the real objection to investing part of the temporary Social Security surplus in non-Treasury securities was that it would reduce the amount of money available for the federal government to spend by an equivalent amount (because money invested in Treasury securities is immediately spent by the government), forcing the government to increase its fiscal deficit, restrain spending, or raise taxes. Today, of course, few seem to be worried about increasing the federal deficit, so perhaps now wouldn't be a bad time for the government to invest part of the Trust Fund in stocks.