Diagnosing Liquidity Addiction

Over the last few weeks markets have recovered from the significant stresses that were building towards the end of May (until yesterday's slow realization). The recovery has been in no small part due to expectations of intervention and that fresh rounds of QE and their equivalents will soon be implemented around the developed world. Deutsche Bank believes that markets are now addicted to stimulus and can’t function properly without it as they show that the periods between central bank balance sheet activity have actually been fairly poor/average periods for risk assets over the last three years.

There is little evidence yet to suggest that markets in this post crisis world have the ability to prosper in a period without heavy intervention, though empirically asset prices benefit from liquidity but that the environment remains fragile enough for them to struggle to maintain their levels when the liquidity stops. Finally, they note that while QE in the US was partly implemented to bring long-term yields down to encourage investors into riskier assets and help lower borrowing/funding rates, the evidence is actually contrary to this.

Critically, they agree with us that the structural problems the West faces mean that QE and its equivalents and refinements will likely need to be around for several years to come to ensure that the financial system and its economies don’t relapse into a depressionary tail-spin. There is no evidence that we are currently close to being able to wean ourselves off our liquidity addiction. The hope would be that with further injections we can prevent the worst case scenario but the base case remains for the stress and intervention cycle repeating itself as far as the eye can see. Central banks still have much to do.


Deutsche Bank: A World Addicted To Liquidity

Risk assets before, during and after monetary stimulus in the US

Here we focus on the performance of risk assets through QE1 and QE2 as well as through Operation Twist. For each program we look not just at the actual announcement dates and program start and finish dates but also the date when it might be argued that further stimulus started to be priced in. Below in Figure 1 we provide a brief description for each of the three programs along with the relevant dates and explanations for those dates.

The Fed’s balance sheet

In Figure 2, we first take a look at the Fed’s balance sheet growth since the crisis started.

The shaded areas highlight the three phases of monetary stimulus that have taken place and we have split each of them into three sections, as described below.

  • The period where we have argued that QE/stimulus started to be expected until it was actually announced.
  • The period from the announcement date to the actual start date.
  • The period from the start date to the end date.

The first major step change occurred in the weeks immediately after the Lehman default where the Fed looked to provide short-term funding to the market as interbank lending ground to a halt. These facilities included the Primary Dealer Credit Facility, the Term Auction Facility, the foreign-exchange swaps with other central banks, the Commercial Paper Funding Facility and the various money market support facilities. After that, the two rounds of QE did see the balance sheet increase by several hundred Billion Dollars. However since QE2 officially ended on June 30th 2011 the Fed’s balance sheet has been broadly static which is understandable given that Operation Twist simply extends the duration of their balance sheet rather than increasing it.


Equities and credit performance

In Figure 3 and Figure 4 we look at the performance of equities and credit through each of the three distinct monetary stimulus programs. Highlighting these phases of stimulus as described above.

The take away from all three programs is that both credit and equities definitely seemed to benefit from the stimulus as in general risk assets had been quite weak leading up to the stimulus before generally performing through much of the stimulus period.

It’s also probably worth noting that the magnitude of the performance seems to have diminished with each round of QE/stimulus. Of perhaps more interest is that since QE2 ended almost exactly a year ago, the S&P 500 has essentially been flat. One would have to say that balance sheet expansion has been more risk positive than simply Twisting.


Commodities – Benefitting from balance sheet expansion but not from Twist?

The story for commodities is fairly interesting. As we can see in Figure 5, focusing on the CRB index, commodities certainly seemed to benefit from the liquidity boost provided by both QE1 and QE2.

It’s also interesting to note towards the end of both QE1 and QE2 commodities started to weaken quite aggressively perhaps indicating their correlation to actual injections of liquidity. Indeed the period around Operation Twist has seen the CRB index fall by around 20%.

Perhaps this is not entirely surprising. If our hypothesis about liquidity being the main reason for the rally during QE1 and QE2 then the fact that Operation Twist didn’t actually add any more liquidity could be a key reason for the lack of positive momentum for commodities.


Treasuries – A confusing picture

With regards to 10 year Treasuries the picture is slightly more confusing. The original broad intention of QE was to bring down yields to encourage money into riskier assets and investments. Figure 6 shows that QE1 actually saw yields rise sharply from around 2% as speculation of QE started to 2.5% on the announcement to 4% as the first round ended. This perhaps shows that the market believed that QE was very positive for the economy which outweighed the reduction of supply of Treasuries in the market place.

Like with QE1, QE2 actually sent yields higher again to around 3.75% within 6 months as hope again prevailed that QE could restore health to the economy. However the data turned in early 2011 and yields fell back to around 3% by the time QE2 ended. Immediately after QE2 ended we then saw 10 year yields rally to below 1.65% in less than 3 months which repeated the extreme rally seen after QE1 ended.

This is pretty much where yields are today as Operation Twist hasn’t had any lasting impact on yields. So overall, although QE was supposed to lower yields, the two largest rallies of the last 3 years have occurred in the period between QE1 and QE2 and then the period between QE2 and Operation Twist.


The US economy before, during and after monetary stimulus

We now turn our attention to how the US economy has reacted to the various stimulus programs. In order to have a fairly timely indicator of economic activity we have used the ISM manufacturing PMI, showing its progression through the various stages of QE.

Focusing initially on QE1 we can see that the US economy saw a strong recovery based on the ISM as it rose from below 35 to around the 60 level. Where QE1 was very successful was that it pulled the economy out of a potential depressionary tailspin. For both QE2 and Operation Twist although the ISM has remained above the important 50 level throughout we have not seen the same kind of improvement in the PMI. In fact during QE2 it actually fell from around 60 to the low 50s where it broadly stands today.

Indeed if we look back through history this recovery is one of the weakest on record in spite of all the Fed actions, plus the three largest peacetime deficits in the US on record. Figure 8 reminds us that only the recovery of 1927 (that ran into the 1929 stock market crash) has been weaker than this one in Nominal GDP terms. This probably tells us that a) the structural problems that encouraged QE were much larger than most believed at the time and b) that QE has limited power to actually power growth forward in such an environment.

With regards to the European economy, it’s also interesting that the PMIs did pick up over the period of the LTROs before falling again immediately after their completion. Europe more than anyone is perhaps addicted and in need of constant intervention to prosper.