Three Charts in a Tangle: What Lending?

UPDATE: Credit where due category: H/T to Felix Salmon for pointing to this chart at bignose.org, which gives you borrowings relative to GDP.

I often hate line graphs, especially when they include outliers that skew the axis, making it virtually unreadable. (This may be a result of seeing several papers this semester with Chilean/US$ exchange rates on a graphic from ca. 1970 forward.)

But this objection is because the historic outlier smoothes the view of current data. It doesn’t apply when you’re living through the outliers. And now that NBER has called the start of the recession, let’s look at some of the evidence. The following three graphics are of two serieses from the Fed’s H.3 Aggregate Reserves data, available from FRED.

First, let’s look at the longest period possible, January 1929*-November 2007:

The Very Noticeable post-WW II increase of the Blue Line (Excess Reserves) is September, 2001. I think we can generally agree that the two-week period in the middle of the month during which very little shopping was done may have had something to do with it.

The rather high (per this graphic, defined loosely as > US$2B) Depository Borrowings are generally May to November of 1984 and April to November of 1988. Whether these have anything to do with the election is left as an exercise. (The pattern is not repeated in 1992 or later.) There are other noticeable high periods—basically, the last months of Richard Nixon’s Presidency (May-Sep 1974) and a few months scattered from October 1979 to June of 1981, and a couple of subsequent outliers, but the Borrowings are generally below US$2B for the month until December of 2007.

Now let’s look at the current century, January 2001-October 2008:

What was a noticeable Crisis Point on the previous graph is now a small bump in the road to Reserve Hell. The scale for Borrowings has gone from peaking at US$9B to US$700B. The scale of Excess Reserves has gone from peaking around US$20B to US$300B—all during the period since the Recession started.

And, just because I’m a cruel person, let’s zoom in on the Recession itself: the past thirteen months of data, October 2007 – October 2008:

The scariest thing about this graphic is the reason it’s on two axes. Note that it is even clearer from this graphic the borrowings from the Fed are far in excess of the Excess Reserves. (Don’t let the narrowing fool you; the slope of the scale of the Borrowings is condensed much more than the scale of the Reserves.

In fact, if we look at the data, December 2007 is the point at which Borrowings first exceeds Reserves. A semi-helpful graphic of the difference:

For October and November, there are over US$1B more Reserves than there are Borrowings. We would be able to see that, except that in December of 2007 there are US$13B more in Borrowings, by April of 2008 the number exceeds US$100B, and it has basically been increasing ever since.**

There may be a few institutions out there lending, but they appear to be doing it primarily with borrowed funds.

*Month chosen because that is where the Excess Reserves data starts being measured. Depository Borrowings starts in 1919.

**There is a slight decline from June (ca. 169) to July (ca. 163.7), possibly due to tax rebate processing; by September, the excess is over US$200B and it soared in October to over US$380B.