For bankers earning slim yields on stagnant pools of loans, higher interest rates might seem like a welcome prospect.
Net interest margins are complex beasts, however, and it is no sure thing that policy tightening by the Federal Reserve would cause them to snap wider. (The following graphic shows data on historical rates and margins, and the relationship between the curve and margins. Interactive controls are described in the captions. Text continues below.)
Rates on 10-year Treasuries fluttered higher in May, perhaps reflecting market expectations that the central bank would taper its asset purchases sooner than previously thought.
While the average yield on the 10-year was higher over the course of the fourth quarter than it was during the third quarter, and it climbed higher still to 1.95% during the first quarter this year, the industrywide yield on earning assets continued to drop. It fell 11 basis points in the fourth quarter and another 11 basis points in the first quarter to 3.72%, according to the Federal Deposit Insurance Corp.
With short rates pinned near zero by the Fed since late 2008, all the action on the yield curve lately has been driven by movements further out on the maturity spectrum.
Data over the last decade or so hints that there is a positive link between how steep the yield curve is and how wide net interest margins are. Such a relationship would seem to hold to the extent that banks lend long and borrow short, though imbalances in the durations of assets and liabilities are risky.
Steeper yield curves, as measured by the difference in the yields on 10-year and 3-month Treasuries, have often been associated with higher industrywide net interest margins during the 46 quarters since the fourth quarter of 2001. The relationship is weak, however, between the curve and the median net interest margin for a group of more than 600 banks whose shares are listed on public exchanges.
There is simply a lot of noise in net interest income figures. The publicly traded median has moved in a tighter range and, notably, the industrywide aggregate was whipsawed by a temporary boost to yields at a few mammoth credit card issuers that adopted new accounting rules in the first quarter of 2010.
Broadly, over a two-decade horizon that spans two recessions, net interest margins have ground down as lending has moved off bank balance sheets. The industrywide margin fell 75 basis points from 1992 to 3.42% in 2012.
An important factor for short-term changes in margins is whether banks are positioned to be hurt or helped by a simultaneous move in rates across the curve.
Regulatory data on amounts of assets that would mature or reprice within a year minus liabilities with the same shelf life can be aggregated across the industry. More short-term assets suggest greater asset sensitivity, meaning yields could rise faster than funding costs, which would improve margins. But the measure is deeply flawed, because it does not capture derivatives, for example.
With a median ratio of net short-term assets to total assets of 5.6% at yearend, top-tier holding companies appear to be a bit more asset sensitive than they were when the Fed last began tightening in 2004 the figure was 5.4% in the third quarter that year.
During that cycle, the central bank lifted short-term rates by about 4 percentage points to about 5% by 2006, long-term rates were comparatively stable, the yield curve turned negative, and margins generally narrowed.
This time is different than the mid-2000s, though. The Fed has deployed unprecedented maneuvers to pull down long-term rates, and margins have dropped to pretty low levels. Besides, higher rates could mean the Fed is tightening because loan growth sluggish for so long is quickening. That could be more important for revenue growth than margins.