QE Not Working Out as Planned, but Hey, I'm Just a Caveman Economist

By Vince Foster  JUL 08, 2013 11:07 AM

The Fed and pundits are declaring QE victory based on employment and rising stock prices, but correlation does not mean causation.

 


With QE tapering now front and center on the minds of the Street’s investment strategists, the analysis has now shifted from whether or not the Fed will taper to when tapering will begin, by how much, and what it could do to the market.  In an effort to get ahead of this tapering event, this past week FTN’s Chris Low, one of the Street’s most well-respected economists, explained QE’s market influence in three bullets:

The three ways QE affected rates. 1. Removed bonds from the market (flow). 2. Relationship between size of portfolio and rates (stock).  3. “Bernanke  put.”  The three ways QE affected rates.

Now I’m really confused.  His number 1 is 2 and his 3 is 1.  After I read this research with the backdrop of the massacre underway in the bond market, I seriously wondered whether anyone really knows how QE works, and more importantly, how it will un-work. 

The debate among academic and investor circles is broken down into a basic stock versus flow argument.  Bond yields are influenced by the size of the Fed’s holdings relative to outstanding supply (stock) and by the process of bidding for securities on the open market (flow).

In a 2011 paper published by the St. Louis Federal Reserve titled Flow and Stock Effects of Large-Scale Treasury Purchases that studied the impact of the QE I program, the authors break down the definition:

“Flow effects” are defined as the response of prices to the ongoing purchase operations and could reflect, on top of portfolio rebalancing activity due to the outcome of the purchases, impairments in liquidity and functioning of the Treasury market…. 

Meanwhile, “stock effects” are defined as persistent changes in price that result from movements along the Treasury demand curve and include the market reaction to changes in expectations about future withdraws of supply.

Concluding: 

We find that both types of effect were statistically and economically significant. Specifically, we estimate that the average purchase operation temporarily reduced yields by about 3.5 basis points and that the program as a whole shifted the yield curve down by up to 30 basis points, with both effects concentrated at short to medium maturities.

Princeton Economist and New York Times columnist Paul Krugman agrees, writing on April 19, 2011:

I basically think of asset prices in a Tobin-type stock equilibrium framework (PDF). People make portfolio choices, allocating their wealth among bonds, stocks, etc. Asset prices – including the famous “q” – rise and fall to match these portfolio choices to the actual asset supplies.

On this view asset purchases matter because over time they change the stocks of assets available: by buying long term federal debt, the Fed takes some of that debt off the market, and hence drives up the price of what’s left, reducing interest rates. The flow – the rate of purchases – matters only to the extent that it affects expected returns.

Chairman Bernanke provided support to the stock view responding to a question from MNI’s Steve Beckner in the April 25, 2012 post-FOMC press conference (emphasis mine):

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.

Over the past few months that theory has been completely blown to pieces.  The Fed has done nothing to change the scope of the program yet the bond market is melting in their face.  The Fed’s balance sheet and flow of purchases both continue to increase, yet the market is completely impervious to their influence. This implies that in the bond market, which is the only thing that matters, QE has become irrelevant, which means Fed policy has become irrelevant.

With all of the truly worthless research analyzing when tapering will begin and by how much -- both of which are irrelevant -- the most important analysis that has gone unaddressed is why the idea of tapering elicited such a massive move in interest rates.  Has any credible Wall Street strategist explained why this is happening?  This is a critical to understanding why the QE trade is blowing up.   

Without going into detail about the Fed’s ex post QE cost/benefit analysis policy error, the simple answer lies in the focus of the program.  When the Fed opted to target the MBS market they entered the arcane realm of interest rate volatility.  Mortgage securities are short volatility by virtue of the prepay option the borrower retains, thus MBSs are much more sensitive to swings in interest rates. The so-called negative convexity exhibited by MBSs sees prices rise less as yields fall, and fall more as yields rise.  This dynamic causes leveraged MBS holders to hedge this risk by buying duration when yields fall and selling duration when yields rise.  You can see how this negative feedback can exacerbate moves in the bond market, especially at extremes.

In order to lower home financing costs to spur housing demand, the Fed announced they were going to buy what amounted to the equivalent of nearly 100% of net new supply.  The initial market reaction was to push MBS yields to record lows and spreads over Treasuries collapsed to near zero.  The effect was that a market that is short volatility was put in a very volatile position.  In other words, contrary to QE II that just focused on accommodation in a relatively benign liquid part of the yield curve, in QE III the Fed exchanged policy accommodation for increased bond market price risk.

MBS Spreads



As I warned right after QE III was launched on October 1, 2012 in The Unintended Consequence of Open-Ended QE:

The last element of increased volatility is the fact that the market is loaded with very low long duration coupons.  When the Fed actually launched QE II in November 2010 the Street was locked and loaded to hit Bernanke’s bid and the 10-year yield rose dramatically as the curve steepened on the back of a rising inflation discount.  The 10-year coupon during those months was in the 2.625% range.  Today the 10-year coupon is 1.625% adding to the effective duration and making it more sensitive to a move in interest rates.  If we get a comparable 100 bps rise in the long end of the curve, you are talking about a massacre in market prices.

So you have a market that is short volatility facing a potentially very volatile environment.  There is the nature of the volatility in the incoming data and how the market calibrates the discount of the ultimate size of the QE program.   You have the changing inflation implications in the data’s effect on the size of QE and that impact on the yield curve.  Then there is the reflexive relationship of the negatively convex MBS market and the added inherently volatile low coupon long duration assets.  Taking all of these ingredients together and Bernanke is cooking up a witches’ brew of market volatility.

When the Fed made it clear that tapering was in the offing, I again sounded the alarm that this market price risk was about to rear its ugly head in Welcome to the Dark Side of QE: The Yield Curve Adjustment Process:

What I think gets lost in the discussion about the cost/benefit of QE, and more importantly, how long it has gone on, is the fact that there are billions of potentially very volatile long duration, low coupons on the balance sheets of investment portfolios….  Due to the nature of these low, long duration coupons, the adjustment process is likely to be very chaotic, making last week look like a game of tiddlywinks.

By focusing on the mortgage market the Fed made a grave tactical error.  The benefit to the borrower in lower interest rates is cost to the investor in lower coupons.  Due to negative convexity in the MBS market, this cost/benefit trade-off is exponential.  The benefit of refinancing a few credit-worthy borrowers from 4.5% to 3.5% had the net effect of swapping MBS investors out of 4.0% coupons into 3.0% and 2.5% coupons.  To the passive participant that doesn’t sound like such a big deal, but in the bond market that is flooded with negatively convex negative coupons it is a major risk factor.  In so doing they made the market much more price-sensitive to the inevitable rise in interest rates. 

The Fed viewed this interest rate risk as negligible because by drinking their own Kool-Aid they believed the stock theory would continue to hold interest rates low regardless of how much flow they continued to buy.  Now we know this stock theory is wrong and the market and economic consequences could be severe. I subscribe neither to the stock or flow theory.  I have put forth that it is the QE (inflation) discount that governs bond yields and asset prices.  The QE influence was found in the real interest rate that was a function of the inflation premium.  Real rates dragged nominal yields lower as the inflation premium rose and is now pushing them higher as that inflation premium is being removed. 

Real 10-Year Vs. 10-Year Inflation Breakeven



As I concluded in The QE Carry Trade is Imploding Right Under Everyone’s Noses:

It’s happening right under everyone’s noses, but they can’t see it because they don’t understand how QE impacts market prices. They are focused on the flow of QE and not the discount.  It’s not the flow of purchases that affects risk assets, it’s the inflation discount embedded in the yield curve. The US economy and capital markets have been reliant on the Fed successfully engineering a negative real interest rate regime. By reneging on their commitment to inflation, the negative real rate regime is imploding -- and with it, the QE carry trade they engineered.

This is a huge deal because this meltdown is happening in the face of decelerating growth and there is not a lot of room for error.  The Fed’s QE is helpless if it is unable to generate inflation, and I believe that is what is driving this violent unwind.  In March I warned what would happen once the market took matters into its own hands in Quantitative Easing: The Greatest Con Ever Sold:

When interest rates rise into decelerating nominal GDP, “accidents” can happen.  If the 10-year is topping and we get a swift move toward 2.5-3.0% as nominal GDP is decelerating then the stock market that everyone loves to own could suddenly become very risky.

Why should we assume that the Fed has been successful in lowering interest rates, or that they would be able to hold them down if the market were ready to turn?  Buying $5 billion/day in a market that trades $500 billion isn’t going to cut it. 

There is no denying the improving employment picture despite the quality of jobs being substandard.  However there is also no denying that the growth in consumption is decelerating to the lowest levels of the recovery.  Not only is consumption the largest component to GDP, the growth rate is highly correlated.  The PCE YoY growth rate in Q4 averaged 3.5% and nominal GDP registered 3.5%.  In Q1 PCE averaged 3.1% growth and nominal GDP grew 3.3%.  Thus far the Q2 run rate for PCE is 2.65% potentially pointing to a sub 3.0% nominal growth rate.  As a comp, when the Fed ended QE II in June 2011 the growth rate in consumption was 5.3%.  

PCE YoY – 10-Year Yield Vs. PCE Deflator



When gauging the growth rate in consumption in the face of rising interest rates, you can see we are entering a zone where the cost of money becomes restrictive.  When taking the 10-year less the PCE deflator, there is even a greater deviation from the falling regression line approaching two standard deviations in real terms. This is perhaps the most underappreciated economic development that equity market bulls and QE victors are ignoring.  These economic growth rates can’t handle rising real interest rates.  Nevertheless the Fed and financial punditry are declaring QE victory based on employment and stock prices, yet it’s obvious they don’t even really understand how it works.  If they don’t know how it works how can they claim it’s responsible for improving employment or higher stock prices?  Correlation does not imply causation. Just because the Fed is buying bonds and stocks are rising doesn’t mean the Fed’s buying bonds is responsible for rising stocks.

Economists like Fed staffers, Paul Krugman, and Chairman Bernanke who subscribe to the stock theory must think a 30 basis point shift in the yield curve as the result of QE is statistically and economically significant.  I’m not sure about that, but what I am sure is statistically and economically significant is a negative convexity blowout that pushes the curve 100 bps higher in  months despite no change in policy. 

But hey, I’m just a caveman.  Their world frightens and confuses me. Sometimes I see the IS/LM curve and think, who is drawing these strange lines, and ask, what does it mean?  I don’t know.  My primitive mind can’t grasp these concepts.  But there is one thing I do know: From the Fed, talk is cheap, and to the market the curve is steep.  And at the end of the day, the only thing that matters is whether the market’s balance sheet can withstand a convexity blowout of a few hundred beeps.

Twitter: @exantefactor
No positions in stocks mentioned.