Banks and savings and loans were allowed to issue NOW accounts nationwide in 1981. An unexpected result was that the relationship between the growth of M1 and GDP broke down. The reason was that customers of depository institutions transferred funds from savings accounts, which are part of M2, into interest-bearing NOW accounts, which are part of M1. The Fed stopped using the growth of M1 as a major policy variable for monetary policy in 1982. Shortly thereafter in 1987, the Fed no longer set growth targets for M1.
In the early 1990s, the growth rate of M2 and a fairly strong relationship between it and GDP began to break down. What happened is that some holders of savings and time deposits moved those funds into mutual funds. M2 decreased but economic activity did not. The upshot was that the Fed could not rely on the past historical relationship between M2 and GDP.
If the relationship between the money supply and the level of economic activity is not predictable using either M1 or M2, what can you say about how well the quantity theory of money predicts the effect of an increase in the money supply on the level of GDP during a recession? What do you think is the explanation?