When the Federal Open Market Committe (or FOMC) meets on Wednesday, the markets will be focused on one thing, and that is whether Bernanke launches another round of quantitative easing (or QE) or extends the maturity extension program dubbed Operation Twist. This past week, the crack team research staffs at various Wall Street investment banks have all chimed in with their respective guesses about what the Fed is likely to do, and the financial headlines have played to the suspense.
Bloomberg ran a story titled "Fed Forecasts Key to Further Easing as Job Market Stalls"
that cited Bernanke’s Joint Economic Committee testimony where he said that “the central question” is whether growth is fast enough to make “material progress” in reducing unemployment.
For the Fed to rely on their internal forecasts
is not only laughable but scary. Since the FOMC’s first economic forecast was published last year, each time growth has been revised down. It’s been so off that if we were conspiracy theorists we might conclude they were intentionally sandbagging the forecast so when we don’t hit the number it will give them more ammo to ease further down the road. It is ironic when you consider their efforts to stimulate are the very reason their forecasts are so awful.
The article also cited many firms including JPMorgan
(JPM), Goldman Sach
s (GS), and Morgan Stanley
(MS), which had all reduced their estimates for GDP. In addition, Bloomberg published a guest commentary by Nomura’s Aichi Amemiya who cited various “key metrics” that showed the odds of additional QE are increasing, including its USD Surprise Index, the Fed’s 5-year forward breakeven inflation rate, and corporate bond spreads.
This week, I didn’t want to dwell on what everyone else has already said but rather posit what might happen when the Fed actually walks away from the bond market.
It’s not that logic needs to apply here, but let’s just think for a moment about what the Fed is contemplating. Like my firm said last week
, the Fed has been actively trying to calibrate aggregate demand via the manipulation of interest rates for over a decade and all we have to show for it is a market that prices every asset like a commodity. So after two years of QE I, QE II, and Operation Twist having produced no net additional stimulus, the Fed is debating about whether or not to continue to more of the same. The definition of insanity
certainly comes to mind.
Over the past two years (March 2010 to March 2012), nominal GDP, which is what needs to accelerate in order to bring down unemployment at a faster rate, has grown at an average of 4.00% on a year-on-year (or YOY) basis. To give you an idea of how that stacks up when they weren’t active in the bond market, the preceding 10 year average (March 2000 to March 2010) growth rate for YOY nominal GDP was 4.10%. Here’s a thought: Maybe the structural growth rate of the economy is 4.00% regardless of whether the Fed is stimulating or not. In fact, you can argue with a high degree of certainty that the quarterly downticks in growth are a direct result of the Fed’s stimulative actions on inflation, which is reducing real consumption.
Correct me if I’m wrong, but I haven’t heard any business leaders announce hiring plans because their input costs are rising but that is exactly what the Fed is trying to do to combat unemployment. I would conclude that the Fed is the single biggest contributor to slowing real growth and stubbornly high joblessness. The fact of the matter is that the economy is going to grow at a top-line structural growth rate no matter what the Fed does and the delta of how that translates into real growth is higher inflation, which the Fed is actively trying to generate.
In regards to Nomura’s “key metrics,” these are similar to other financial conditions indices including the St. Louis Fed’s Financial Stress Index
in which one of the “key” denominators is the 10 year (or 10YR) yield. That’s right -- risk premiums that look to be blowing out to the upside are spread versus the 10YR. So the Fed -- by actively trying to reduce the 10YR yield to stimulate borrowing -- is in effect raising the risk premiums that are causing financial stress. If you look at nominal corporate and equity yields, they are not rising much with the vast majority of the spread widening occurring with the 10YR.
In fact, if you use a normalized benchmark for the risk-free rate in nominal GDP growth and spread that over corporate and equity yields, the risk premiums are actually tightening and the credit risk premium is 200 basis points ("bps") under the 30 year historical average. It is not exactly a 2008 scenario where spreads were blowing out due to corporate and earnings yields rising. Risk premiums aren’t reflecting increased credit risk; they are reflecting increased interest rate risk and for good reason.
In the fall of 2011, I was watching an interview on Bloomberg Television with Tom McClellan
of McClellan Oscillator fame. Tom had identified a remarkable correlation with the S&P 500
(^GSPC) and the CFTC Commitment of Traders net position of commercial hedgers in eurodollar contracts. Poor Adam Johnson kept thinking McClellan was referring to EUR futures. But having traded “euros” while working on the street, I knew he was talking about the 90 day LIBOR kind.
For those of you who don’t know, the eurodollar pit at the CME is the most liquid marketplace in the world. Each euro contract has a notional value of $1 million and will trade close to two million contracts daily equivalent to $2 trillion in notional value. The commercial hedgers are, of course, the global money center banks. McClellan had figured out that the net position of the commercial hedgers was highly correlated with the S&P with a one year lag. I was intrigued by this statement and tried to replicate it on my Bloomberg terminal the following day. It was indeed a remarkable correlation and not only called the March 2009 bottom but also the 2010 high and low, the 2011 high and low, and the 2012 high with a low in June, which we possibly hit at 1265.
McClellan was not clear on why there was a correlation but I have a possible thesis that makes sense. Commercial banks would buy euro contracts if they needed to hedge against falling
LIBOR, which they would do if they owned a floating rate asset because their interest rate exposure was to the downside. In October 2010, the commercial hedgers’ net position was short 1.171 million contracts. But a year later in November 2011, that had reversed to long 1.149 million contracts for a notional equivalent swing of $2.32 trillion.
In other words, it seems as if the banking community felt compelled to hedge $2 trillion in floating rate assets over a 13 month period. Now was that new credit that was extended into the economy? I can’t be certain, but it would make sense that the hedging need of the banking system correlates to the equity market with a 12 month lag since companies would take that capital and deploy it to earn a return, which then leads to capital growth over the subsequent year thus showing up in higher stock prices.
There are two financial metrics on the S&P 500 that have a tight correlation with loan growth. These are Return on Assets ("ROA") and Return on Capital ("ROC"). Currently the S&P 500 ROA is near a decade high at 9.33%. When companies can earn high ROA, they tend to borrow more capital to buy more earning assets. The YOY growth in Commercial and Industrial (C&I) loans at last print was 16% and has been steadily near the 10% area since the beginning of 2011. The trajectory of C&I loan growth is, not surprisingly, positively correlated with the ROA of the S&P 500. In addition, the S&P 500 ROC less the cost of capital (measured by Moody’s BAA bond yield) is at 11% also near decade highs. The S&P 500 ROC – cost of capital is also positively correlated with the S&P 500 price, which makes sense and supports the correlation versus the eurodollar commercial hedger position.
In other words, when companies earn positive returns on assets, they tend to borrow capital to invest in more assets; when they earn a positive return on that capital less the cost of that capital, that translates into higher stock prices.
We can go a step further and look at other relationships that support this conclusion. Upon inspecting the eurodollar commercial hedgers net position, I wanted to drill down on how this showed up in the price of LIBOR during the period of increased hedging. If you overlay the net position over 90 day LIBOR, you will find that it is positively correlated. This means that as more hedging took place, LIBOR actually rose. This makes sense since as the demand for money increased, the cost of money increased as well. Recall during the August crash last year, LIBOR increased from 25bps to 50bps in a six month period. Many pointed to this as suggesting the financial system was in distress but looking at it over the bank’s hedging needs, it looks like they were actually substantially increasing their loan books as we noted above. At the same time, what seemed to follow increased loan hedging and increased LIBOR rates was an increase in the value of the dollar as measured by the DXY. While many attribute the dollar rally to weakness in Europe and a flight to quality, a different interpretation could be that the US economy is absorbing the excess and idle liquidity on bank balance sheets, which is net/net dollar bullish.
We’ve discussed what the Fed is not doing in stimulating economic growth via QE but let’s now discuss what they are doing in re-risking the banking system.
Banks have historically low loan/deposit ratios near 80%. As a consequence, they have very high concentrations of securities on their balance sheet. Banks tend to traffic, not in bullet US Treasuries, but in callable bonds such as agencies and mortgage backed securities. With the Fed pushing yields to historic lows, many banks are seeing previously bought bonds called and are being forced to reinvest the cash in lower coupon callable bonds, which reduces net interest margin while increasing interest rate and balance sheet risk.
The Fed likes to extol the benefit of low interest rates on borrowers without discussing the costs to the lender/investor and it’s not just in lower yield. The lower the coupon for a given duration, the more volatile that coupon becomes. Currently (with virtually the entire curve sporting negative coupons) there is tremendous embedded volatility on bank balance sheets. Add to that the fact that the banks' holdings are primarily callable and thus short convexity, and you have the recipe for a very unpleasant situation when rates start to rise.
Convexity can be thought of as the second derivative of duration. Bonds that are positively convex rise more in price when yields fall and fall less in price when yields rise. Conversely, bonds that are negatively convex rise less in price when yields fall and more in price when yields rise. This is also known as extension risk: When rates rise, what was once a short callable becomes a long bullet increasing the duration, and thus mark to market risk on a bank’s balance sheet. Of course, this does not have to be a problem if a bank is willing to hold the bond to maturity, but if they need to sell bonds to fund loans, banks will be in a precarious situation of having to realize the loss on their portfolio or otherwise increase their balance sheet via borrowings. In other words, a sell-off in the bond market would make the highly liquid banking sector very illiquid very rapidly.
Recently Stanford economist John Taylor (Taylor Rule) wrote on his Economics One
blog an entry about how much of net new borrowing by the US Treasury had been bought by the Fed in 2011. Using Government Printing Office
data (that’s not a joke), he calculated that the Fed bought the equivalent of 77% of the net increase in the Federal debt in 2011, calling it an “amazing percentage.” This is an interesting data point as we approach the end of Operation Twist and one we have raised in the past in the stock versus flow debate
During the last FOMC press conference, Bernanke was asked by MNI’s Steven Beckner whether he thought there would be a market reaction when the Fed ceased purchases. Bernanke
gave the Fed’s stock thesis saying:
There’s some disagreement, I think, about exactly how balance sheet actions by the Federal Reserve affect Treasury yields and other asset prices. The view that we have generally taken at the Fed in which I think – for which I think the evidence is pretty good is that it’s the quantity of securities held by the Fed at a given time, rather than the new purchases, the flow of new purchases, which is the primary determinant of interest rates. And if that is – if that theory is correct, then at such time that our purchases come to an end, there should be relatively minimal effects on interest rates at that time.
That’s a pretty bold statement coming from the guy who has been the single largest source of demand in the bond market and especially now that he has loaded the banking system full of very volatile interest rate risk. He might be right, but he might be wrong and we know that Bill Gross, who owns a few bonds himself, subscribes to the flow theory among others.
The truth is nobody really knows. But what we do know is that the US Treasury has doubled the supply of the float in a very short period of time while the Fed has forced the coupons to the lowest in history, providing a very high false sense of natural demand in the market. The Fed has basically engineered a blow-off short squeeze... but as traders know, when a squeeze plays itself out, price typically reverts back to the natural balance of supply and demand, which I believe is at much higher yields, European meltdown or not.
Speaking of European meltdown, one of the more interesting aspects of the crisis is the interplay between currency, credit, and interest rate risk. By all accounts the debt crisis stems from current account imbalances between the northern exporting countries and the southern importing countries.
Due to the currency union, the southern countries were provided more purchasing power than their domestic currencies would have allowed, which led them to consume more than they otherwise should. This led to excessive debt levels relative to the size of the economy. What needs to happen is a broad based devaluation among EMU members to adjust purchasing power -- but because they all share the same currency, the market is not being allowed to function properly so it is forcing a devaluation via the sovereign bonds. You can think of current debt prices as proxies for devalued currencies so you have essentially transferred what should be interest rate risk (inflation via currency devaluation) into credit risk (ability to service the debt).
This is not too dissimilar to what has occurred in the US parallel universe, which is in a similar but inverse credit position -- only we have the ability to print dollars and the market is allowed to devalue the currency. If the US were on the gold standard or the dollar were pegged to the quasi German mark, like the PIIGS, we’d be in the same position. The question for the bond market is: When does US credit risk that has been inflated via currency devaluation become interest rate risk?
Ironically, by transferring private sector credit risk to public sector credit risk, the Fed has generated a massive amount of interest rate risk in the US banking system. The longer they wait to exit their stimulus, the more negative-coupon negative-convex long duration assets get loaded onto bank balance sheets and the more painful the eventual unwind will be.