‘QE is useless!’ Phrases like that have made the headlines on more than one occasion. Many agree that Quantitative Easing, the monetary bazooka used by most central banks to overcome the financial crises, has not aided the economy, while asset bubbles have emerged almost everywhere. Sounds familiar, right? But is it in agreement with reality?
Pushing down yields
As central banks have explicitly or implicitly pointed out, the primary goal for using QE measures is to push interest rates down. This implies that, assuming QE is effective, the more easing is deployed the lower the interest rate. To test this hypothesis, I plot the ‘amount’ of QE against the change in the interest rate for the ‘big four’;
- the Federal Reserve (Fed),
- the Bank of Japan (BoJ),
- the European Central Bank (ECB) and
- the Bank of England (BoE).
The amount of QE is defined as the (percentage) increase in the size of the central bank balance sheet, while the change in interest rates is based on the 10-year government bond yield (for the ECB I take the German 10-year bond yield). The starting point of this analysis is September 2008. This might seem arbitrary, but it is not considering that central banks globally joined forces and expanded their balance sheets for the first time in October 2008, right after Lehman fell.
The scatter graph below shows the amount of QE (x-axis) plotted against the change in yields (y-axis). The results are surprising. The graph shows that the relationship between the two factors is the exact opposite as assumed above. Less QE is accompanied by bigger yield changes, and vice versa. The ECB needed just 16% of balance sheet expansion to realize a 3% lower yield. The Fed, on the other hand, blew up its balance sheet by a massive 200%, but did not manage to push the yield down by more than 1.3%. The relationship between QE and yield change is highly significant. The ‘best-fitting’ straight line, which describes the relationship, is able to explain 76% of the variation of the four data points. I should add, though, that a low number of data points artificially drives up the explanatory power (two data points will form a perfect relationship).
Click to Enlarge
The hypothesis that more QE leads to bigger changes in the interest rate is rejected. Perhaps this is related to the fact that the Fed already started lowering short-term rates in 2007, while it took the ECB until August 2008 to cut interest rates for the first time. Also, in recent months the Fed is heading for the ‘QE-exit’, where the ECB looks ready to start another round of easing. This does not take away, however, that the relationship between QE and interest rate is highly significant.
What about equity markets? To get an impression of the effect on stocks I plot the amount of QE against the total return on the S&P500, the Nikkei, the Stoxx 50 and the FTSE 100. The results are shown in the graph below. Again, one picture is worth a thousand words. The more QE, the higher the realized return. For the US a 200% balance sheet expansion yielded return of 100%, while in Europe 16% of QE only got you 20% total return. As can be derived from the graph, a line fits the four points almost perfectly, explaining 94% of the variation. For those who support the ‘central bank bubble blowing’ theory this chart is probably much appreciated.
Click to Enlarge
But, portraying central banks as a bunch of bubble blowers is just a bit too easy. Assumptions that QE has no meaningful impact on the economy are premature. To understand why, please look at the third scatter plot shown below. The amount of QE is again plotted on the x-axis, but this time the y-axis shows the realized GDP growth of the four regions (for the ECB I take the Euro zone GDP numbers).
Click to Enlarge
More QE = More GDP
This is a pretty neat way to demonstrate that QE is not only related with rising stock markets, but also with economic growth. The scatter plot reveals that growth has been the strongest in the region where its central bank eased the most. The US economy has grown 6% since the third quarter of 2008, while the Euro zone economy actually shrunk as the ECB refrained from heavy and long-lasting QE stimulus. In addition, the relationship between QE and GDP growth is the strongest from a statistical viewpoint. With an explanatory power of 97%, it’s almost perfectly linear. Roughly 25% of balance sheet expansion is accompanied by 1% GDP growth. I wonder if the ECB is aware of this chart.
The three scatter plots show that QE does matter. Differences in the amount of QE by central banks are able to explain large parts of the differences in interest rate movements, stock market performances and GDP growth. Especially that last finding is different from what is often perceived. Now, admittedly, there are many possible explanations for these findings, like the more flexible nature of the US economy and perhaps the Fed’s monetary policy, or the fact Europe had to deal with two crises (peripheral debt issue), and so on. But to state that QE does not have any (economical) impact, is just too easy.