Loading...
Most Recent Stories

The Myth of “Quantitative Tightening”

Over the last 7 years I have repeatedly debunked many of the myths associated with Quantitative Easing.  I said the policy probably wouldn’t have much impact on the economy, wouldn’t cause interest rates to spike, wouldn’t cause high inflation, shouldn’t be referred properly as “money printing” , wouldn’t lead banks to “multiply” their reserves and could even be marginally deflationary.  Many of these views have since been adopted by major Central Banks, pundits and mainstream economists.  I’d say I have a decent track record explaining how QE actually works and how its various transmission mechanisms might impact the broader economy.  Unfortunately, the myths about QE continue to this day.

One trend I’ve started to notice in the last few months is the idea of “Quantitative Tightening”.  This is the idea that a Central Bank that unwinds its expanded balance sheet must necessarily have a contractionary effect on the economy. This concept has been referred to recently in the case of China’s Central Bank and some of their sales of US Government bonds.

This is really two different concepts.  The first is a domestic Central Bank’s QE and the second is a Central Bank’s response to aggregate trade flows.

First, Quantitative “Easing” is a bit of a misnomer.  We know from the many iterations of QE in the USA that this policy isn’t necessarily all that easing.  This is due to its weak transmission mechanism.  QE has six primary transmission mechanisms including the interest rate channel, the financial crisis channel , the asset price effects (wealth effects & housing & equity price channels), expectations channel, credit channel and the exchange rate channel.

Most people think QE is powerful because it adds “money” to the private sector, but as I’ve explained repeatedly QE is really just an asset swap.  The more “money” is only stimulative to the extent that swapping a savings account with a checking account is stimulative. So, we have our other six transmission mechanisms.  The interest rate channel is weak because QE doesn’t have a strong impact on long rates unless the Central Bank explicitly sets long rates.  Otherwise, long rates float with market expectations.  The financial crisis channel works best when markets are panicked and liquidity is low.  The asset price effect is highly questionable and not likely as powerful as some believe.  The expectations channel can’t possibly be as powerful as some assert because of the multi-temporal problem I’ve discussed before (basically, expectations of a boost in growth now is offset by expectations of a contraction in policy later).  The credit channel isn’t powerful because the central bank doesn’t directly control demand for debt.  And the exchange rate channel is only powerful if the central bank explicitly targets exchange rates.  So, the short story is, QE isn’t really as “easing” as many believe.

So, the clear conclusion here is that domestic QT can’t be “tightening” unless you believe that QE is a powerful form of “easing”. The evidence on that is weak at best.

At an international basis the term Quantitative Tightening is equally misleading. For instance, China does not control the quantity of dollars that exist in the financial system at any given time. They can influence their domestic economy’s quantity of Yuan, but they do not control foreign currency quantities.  Remember, China buys US T-Bonds primarily because they are a net importer of US Dollars due to their trade position. The People’s Bank of China does not control this position.  The cause (trade inflows) and effect (PBoC T-Bond purchase) is not properly referred to as “easing” since the PBoC is not implementing this policy in an independent attempt to ease policy. Likewise, any unwind of this balance sheet will not reduce global liquidity, but rather, will likely be implemented as an attempt to boost domestic demand and further positively influence trade inflows into China.

Like Quantitative “Easing”, Quantitative “Tightening” likely isn’t as relevant as many will make it out to be.

Some related work: