Caroline Baum has a nice explanation:
The relevance of the funds rate, at which banks borrow from one another for last-minute funding needs or to satisfy their reserve requirements, has nothing to do with the volume of transactions at that rate. Its importance lies in what it says about the thrust of monetary policy or, to put it another way, about the impetus for the central bank to create money.
When the funds rate is significantly below the long-term rate, the Fed has to create excess money to keep it there. Otherwise, it would rise in response to the same forces -- supply and demand -- pushing up long rates.
Alternatively, when the short rate is above the long rate, the central bank has to withdraw liquidity from the banking system to keep the funds rate from following long rates down.
That's as succinct an exposition as you are likely to find about why some of us worry about flat yield curves. The rest of the column is a little more confusing, as Ms. Baum takes issue with this:
The notion that the funds rate is inconsequential always gains popular appeal at a time when the Fed is embarked on a series of interest-rate changes, the results of which have yet to manifest themselves. The leading purveyors of that idea are the keepers of the flame.
Fed Chairman Bernanke pretty much summed up the official party line in a March 30, 2005, speech, when he was one of seven governors on the board. The Fed has "no direct control over the key interest rates and asset prices that jointly determine the extent of financial stimulus,'' Bernanke said. Instead, policy makers are stuck setting "an otherwise obscure short-term interest rate, the federal funds rate.''
If, as Bernanke said, "Monetary policy is effective only to the extent that Federal Reserve actions can affect a wide range of interest rates and asset prices,'' why choose such a lowly instrument for the policy rate? Why not choose the rate that matters?
I think it incorrect to read the Chairman's remarks as suggesting "the funds rate is inconsequential." The point is that monetary policy operates through an indirect channel that can involve short-run changes in real economic activity, longer-run effects on the rate of inflation, and movements in expectations about both. It is indeed long-term interest rates that matter most, but they have a complicated life, influenced by every conceivable change in the economic environment. Monetary policy is but one piece of the puzzle. Just ask Jim Hamilton.