One neat thing about being an economist is that you're always the life of the party. Fill up a room with some actors, rock stars and a supermodel or two, and after 20 minutes, they're all listening to an impromptu lecture on the J-curve implications of last week's rally in the euro. You get used to it after a while.

Why the fascination? It's really quite simple. While the others have individual life experiences, we economists have data on the entire world. We can create tables, formulas and graphs, and as I told Rod Stewart long ago (OK, I didn't really), "Remember, every picture tells a story."

Such is the case with the brokerage sector. Back in the old days, when people still thought the Federal Reserve had a six-shooter loaded with magic bullets, the assumption was that brokerage stocks benefited from lower interest rates and a steep yield curve. Such credit-market conditions facilitate the underwriting of securities and the financing of portfolios. However, since the general equity-market peak of early September 2000, the 13-member equal-weighted American Stock Exchange Broker/Dealer Index, or XBD, has a total return of negative 36.3%, far below the S&P 500's 21.99% drop.

Of course, monetary policy has not been as loose as you might imagine, given all of this year's rate cuts. We can measure credit conditions at the short end of the yield curve by the forward rate between three months and two years. This is the rate at which we can lock in funding for seven quarters, starting one quarter from now. If the ratio of this forward rate to the two-year rate is greater than 1, then the yield curve is positively sloped. Incredibly, this didn't happen until the Fed's March 20 rate cut -- the third of the year. More important, this measure has reflected less-loose credit conditions ever since the May 18 rate cut. This backsliding coincides with the stock market's retreat since the April-May rally.

Not That Loose in the Short End

Source: Bloomberg

Banks vs. Brokerages

In fairness, the 24-member, capitalization-weighted S&P Bank Composite index, or SPBNKC, has a positive total return of 13.31% since last September, indicating that banks have benefited from the wider margins between their funding costs and the return on their loan portfolios. Bank loan portfolios no doubt have a longer duration than do brokerage security portfolios. The relative performance of banks to brokerages since last September is quite striking; the two groups have a near mirror-image track record over the past year.

Performance: It's All Relative

Source: Bloomberg

A New Relationship

A general theme of market analysis since the start of the Asian crisis more than four years ago is "everything you know is wrong." It's time to toss the textbook out once again, for the long-term relationships between various financial powerhouses and key market indices have changed. It shouldn't shock anyone that such members of the XBD as E*Trade (ET:NYSE - news - commentary), Ameritrade (AMTD:Nasdaq - news - commentary) and Charles Schwab (SCH:NYSE - news - commentary) have been hurt by the end of the bull market. It may, however, come as a surprise that more traditional investment banks and brokerages are tied almost as closely to the tech-heavy Nasdaq 100 as to the S&P 500, which has nearly a 10% weight in various financial groups.

Long-term data analysis in the brokerage sector is virtually impossible. Goldman Sachs (GS:NYSE - news - commentary), Citigroup (C:NYSE - news - commentary), J.P. Morgan Chase (JPM:NYSE - news - commentary) and Morgan Stanley (MWD:NYSE - news - commentary) all have undergone significant corporate actions, such as mergers and divestitures, in the past three years.

What we can do is look at a snapshot of how each of these stocks has fared in relation to the Nasdaq 100 and S&P 500 since last September's peak. For each XBD member, as well as for Citigroup and J.P Morgan Chase, we can compare the relative movement of the stock against the S&P 500 and Nasdaq 100. Two key numbers are provided: the beta, or the slope of a best-fit (regression) line; and the R-squared, or percentage of the stock's variance that can be explained by the index.

A perfect fit would have both a beta and an R-squared of 1. A beta greater than 1 means the stock is more volatile than the index, and a beta less than 1 means the stock is less volatile than the index. In the table below, E*Trade is 3.16 times as volatile as the S&P 500, and the S&P 500 explains only 50% of the variance in E*Trade. The table below is ranked by R-squared to the Nasdaq 100. In terms of the R-squared figure, any number more than 0.50 is high and any greater than 0.80 is a good proxy for an index as a whole.

The Brokers' Relationship to the Indices The beta indicates the stock's volatility relative to the index, and the R-squared number shows how much of the stock's variance can be explained by the index
	S&P 500		Nasdaq 100
	Beta	R-squared	Beta	R-squared
E*Trade	3.16	0.50	1.12	0.49
Goldman Sachs	1.72	0.53	0.56	0.45
Morgan Stanley	2.13	0.54	0.69	0.44
Charles Schwab	2.43	0.53	0.85	0.42
Lehman Brothers	2.02	0.50	0.64	0.40
Merrill Lynch	1.73	0.50	0.53	0.37
A.G. Edwards	1.41	0.49	0.42	0.35
Ameritrade	2.47	0.33	0.89	0.34
TD Waterhouse	1.50	0.41	0.66	0.34
Bear Stearns	1.53	0.44	0.47	0.33
J.P. Morgan Chase	1.56	0.52	0.45	0.33
Citigroup	1.35	0.56	0.35	0.31
Raymond James	1.47	0.40	0.45	0.30
Legg Mason	1.29	0.43	0.38	0.30
Jefferies Group	1.02	0.23	0.34	0.20
Source: Bloomberg

E*Trade has the closest relationship to the Nasdaq 100, but Goldman Sachs and Morgan Stanley are close behind. Moreover, the betas of these investment banks to the S&P 500 are quite high as well. The downturn in trading volume and IPO underwriting has taken its toll on all of Wall Street, not just on the retail brokerages maligned in some quarters for their focus on technology.

The sad moral of this story is that the burst financial bubble exploded over all of us, even those who never had a dollar in the market. It was true for actors, rock stars and supermodels, and that's why they listen so intently.