Safe assets and the QE enigma
“The phrase "safe assets" should really be retired, it is a source of so much confusion. "Safe asset" yields are depressed by the fact that it is the policy of the Federal Reserve to make them depressed. This is what I mean. The story is basically "if people were prepared to extend credit, then credit would expand".” – Dan Davies
Snooping on an exchange between Steve Waldman and Dan Davies this morning about Tyler Cowen’s “entrance fee” theory got me thinking again about the narratives that we have built around QE. In particular the idea that whether QE is inflationary or deflationary is key to the story of what exactly large-scale asset purchases (LSAPs) do has caused tempers to flare but mostly, it seems, added to the general confusion.
I want to start with the basics, as summarised by Dan in the quote at the top of the post. We know that as the sub-prime mortgage crisis hit it created a cascade effect on bank balance sheets forcing widespread deleveraging and consequent economic contraction. We know too that central banks tried to offset this both by dropping interest rates to near zero and bringing in LSAPs.
Cowen’s point is that demand for some of the assets targeted under QE programmes, in particular government securities, has driven real returns negative. As he puts it:
“[F]inancial institutions and traders have to hold large quantities of T-Bills (and similar assets) to participate in financial markets. That may be to satisfy collateral requirements, to meet government regulations, to be credible in private market transactions, and so on….The demand for these assets is now so high and so persistent that the assets have persistently low nominal returns and often negative real returns.”
Cowen, indirectly, points to one of the charges levelled at LSAP programmes – that it decreases the supply and increases the cost of collateral. Of course, as Dan says, it is a deliberate policy aim of central banks to depress yields on government bonds in order to make it more appealing for investors to invest in other riskier assets. Critics, however, argue that this impairs financial markets further and prevents credit expansion into the real economy through rehypothecation (looking at you Richard).
But here I think we need to take a step back and ask why these institutions remain so eager to pawn their assets to central banks in the first place? If genuine investment opportunities existed, credit demand was there and the value of collateral held was rising, would it not make sense for firms to lend into the real economy rather than accept negative rates?
It is understandable that investment opportunities might remain unexploited for some time after a shock on the scale that we saw with the financial crisis. Yet five years on we rely on an assumption that, as Steve puts it, the “fincrisis wiped away a forest of cleverness, and that new innovations to eliminate non-bank liquidity constraints might help”. That’s a big call, and one that I don’t think the evidence so far quite stretches to (yet).
One alternative answer is that QE and the collateral shortage is not to blame for the collapse in credit expansion. Rather if expected risk-adjusted return on investment is strongly negative and interest rates stuck at the zero lower bound cannot fall far enough to encourage firms to extend credit then we could observe a similar result.
Indeed the investment dearth theory was being discussed well before the crisis struck. Ben Bernanke referenced it in his 2005 speech on The Global Saving Glut:
“After the stock-market decline that began in March 2000, new capital investment and thus the demand for financing waned around the world. Yet desired global saving remained strong. The textbook analysis suggests that, with desired saving outstripping desired investment, the real rate of interest should fall to equilibrate the market for global saving.”
This story is one of too much money chasing too few opportunities. So where does QE fit into it?
In effect QE imposes an opportunity cost on individual agents holding onto the targeted assets. Those who sell assets to the Fed can redeploy that capital in, for example, equity markets driving them up and creating a strong incentive for others still holding the assets to jump into the rally. This incentive is even stronger if they are struggling to identify productive areas where capital can be deployed for the longer term. (Again it should be noted that this portfolio balance and wealth effects were intended transmission mechanisms of the policy – though I think it a fair criticism that policymakers overestimated their impact on the real economy)
If the investment dearth theory holds then the reason why the Fed was able to hoover up “safe assets” was the lack of investment opportunities, not the collateral shortage causing an investment shortfall. It is important to stress that this is not to say there are no profitable investments out there, only that there are relatively few that offer good risk-adjusted return profiles.
Under this model those currently in need of quality collateral are not the lenders but the borrowers. It is little surprise then that an improving housing market is providing signs of household credit expansion. An increase in property market activity could have the additional consequence of driving private sector collateral creation, potentially offsetting the reduction in government securities issuance due to planned fiscal consolidation.
So to sum up I suspect Dan Davies might be right that there has been too much focus on collateral shortages as a choke on lending. Of course, the worrying implication of all this is that Larry Summers may also be right – we might now need housing bubbles to maintain our growth.
[Incidentally on the subject of unintended QE impacts, the FT’s Robin Harding points out one amusing possible interpretation of a recent Philadelphia Fed paper is that the programme may have allowed people to retire earlier by raising asset prices leading to some of the observed falls in US labour force participation]