If the money supply increases 10 percent in 3 months, we call it a 40 percent annualized increase. If it flattens at three months and is at the same level 3 months after that, it becomes a 20 percent annual rate for the 6 months. Six months later, if money growth remains flat, it becomes a 10 percent annual rate.
This is an oversimplification, but something like that has been happening. The money supply spiked, and then flattened out. Pre-spike until now still gives us pretty big numbers, but they are getting smaller every day. Yet, commentators treat the money growth statistics as if the rapid rise is ongoing.
I don't know the economic implications of the difference in a steady money growth and an equal percentage money growth achieved by a spike followed by flattening. However, I can't imagine that the latter pattern-the recent pattern-is as stimulative.
Turning to the different subject of how to lie with statistics, I've seen lots of graphs lately that have had the horizontal axis squeezed together to produce scary money-growth graphs. The most egregious is probably the graph of the monetary base growth dramatic enough to make You Tube. Shame!