Interest Rates in Wonderland

"Well, I never heard it before, but it sounds uncommon nonsense"

                                 - The Mock Turtle, Alice in Wonderland

"Well now that we've seen each other," said the Unicorn, "if you'll believe in me, I'll believe in you."

                                 - The Unicorn, Through the Looking-Glass


As negative interest rates strengthen their grip on global policy and markets, Macro Man cannot help but a feel a sense of unease, a visceral disquiet that there is something very wrong with the world.  He suspects that he is not alone in this regard.

Although some of these misgivings no doubt arise because negative interest rates just "feel" wrong, ultimately there are more tangible reasons for concern because of what negative interest rates represent and what the implications are.   Looked at simplistically, negative rates entail a government-sanctioned authority imposing a charge upon private sector agents, with the proceeds helping to either directly (via QE bond purchases) or indirectly (via lower bond yields deriving from private sector buying) provide a subsidy to the central government.

There are two reasons to object to this sort of policy prescription in the current circumstance.   The first is that charging the private sector to provide funds to the public sector is tantamount to a tax, in this case on banks (and, potentially, depositors).  Such a policy would appear to be perverse, given the implicit goals of increasing monetary velocity, credit creation, economic growth, and inflation. 

The obvious rejoinder to this charge is that something that is "essentially" a tax is not in fact officially a tax, and unlike official taxes avoiding and evading this particular tax is allowed, nay encouraged.   Indeed, avoiding the tax through deploying cash in "more productive" pursuits such as lending, riskier investments, and consumption is the entire raison d'etre of the policy.  While this may be true, there is also a secondary consideration: what sort of impact are negative rates likely to have on inflation expectations?

While Macro Man is not aware of any empirical or theoretical study on the subject, it seems to him that a negative nominal return on cash would likely lower, rather than raise, such expectations, insofar as it represents a loss in nominal future spending power across the economy.   If one assumes a constant real spending power (a heroic assumption, perhaps, but no less heroic than many economists' assumptions), then inflation expectations have to come down to match the reduced nominal spending power of cash.

The second objection to negative rates is more troubling from a philosophical perspective.  Central banks are comprised of unelected officials who, with certain exceptions, are completely independent from the governments who appoint them.  Yet these monetary officials are crafting a policy that looks suspiciously like taxation- a fiscal measure that comes exclusively under the remit of elected governments.   If citizens do not like the taxation policies of their governments, they can vote them out.   If they do not like the the effective taxation policies of their central banks, they can...do what?  Particularly in the case of the Eurozone, it is unclear exactly how the populace could dispense with Draghi's services should popular unrest come to that.  (To be clear: Macro Man is not suggesting that it should or will.)

The Swedish Riksbank published a note last year on negative rates that includes a useful schematic, reproduced below.   The upshot is that there is a limit to how negative rates can go, insofar as a) banks would not lower their lending rates in concert with deposit rate cuts, and b) even if banks begin to charge negative rates on deposits, at a certain level depositors would be incentivized to withdraw their money and hold cash, despite the transactional friction of doing so.

                                  Source: http://www.riksbank.se/Documents/Rapporter/Ekonomiska_kommentarer/2015/rap_ek_kom_nr11_150929_eng.pdf

In extremis, political and monetary authorities could drag the the economy fully down the rabbit hole while pushing back against cash withdrawals to ensure that their policies regained/maintained traction.  An obvious way to rekindle a preference for electronic as opposed to physical money would be to implement a so-called "Gesell tax", which entails a tax on cash via a requirement that physical notes be stamped periodically to remain legal tender.   Although this is no more radical than the steps taken in the 1930's- including the ban on private ownership of gold- in the modern political climate it would appear to be a policy ideally conceived to foment revolution.   Indeed, a Stamp Tax on printed materials was one of the key grievances of the American colonists in the decade preceding the Declaration of Independence.

As of yet, negative interest rates are still in their infancy, so the empirical evidence as to their impact upon the real economy is fairly scant.  To be sure, we can observe price action in government bond markets, but in the case of the Eurozone and Sweden there is also the issue of central bank purchases to consider, and it is difficult to distinguish between the two as to impact on bond markets.

However, one of the criticisms of NIRP is that it will adversely impact banks' margins, and that is something we can measure.  The chart below is an adaptation of one from the Riksbank paper linked above.  It shows the Riksbank deposit rate, the rate on demand deposits (in this case, those held by households, although non-financial corporation deposit rates are virtually identical), the composite loan rate on the stock of existing bank loans to households and NFCs, and the spread between the loan and deposit rates.  As you can see, margins have plunged to the lowest level in the history of the dataset, although they are really just continuing a trend set in motion when Riksbank deposit rates went to zero.

The phenomenon is more clearly observable in the Eurozone.   Using a similar dataset (though in this case, the loan rate is a composite cost of borrowing for NFCs), we can clearly observe a clear decline in both loan rates and margins since the imposition of NIRP in June 2014.  Note that these charts do not account for the negative margin between the rate on banks' consumer deposit liabilities and central bank deposit facility assets.

To be fair, this has coincided with a long-awaited increase in MFI lending to the private sector, which was one of the points of the whole exercise, so it is difficult to call it a failure at this juncture.  By the same token, if margin compression throws banks back into the mire, as it appears to have done recently, the renaissance in the flow of credit may prove ephemeral.   It is also important to recognize that the regulatory burden imposed by loans also probably limits the utility of monetary policy to kindle banks' desire to lend.

As for the impact upon the real economy, it is difficult to draw any firm conclusions given the dearth of available samples and our inability to conduct proper scientific experiments.  (In other words, there is no way to empirically test counterfactual scenarios as there is in a controlled laboratory environment.)  Nevertheless, Macro Man sifted through the scraps of what little data we have to try and identify any trends in the real economy.

He looked at Denmark and Sweden (time constraints and an accidentally-deleted spreadsheet precluded the inclusion of the Eurozone in this piece, unfortunately) and attempted to judge the change in certain domestically-sensitive economic variables: nominal wages, house prices, and the relative performance of bank stocks to the broader equity index.   He acknowledges that this is an imperfect methodology, particularly as the wage and house price data are only released quarterly; nevertheless, we may be able to distill some insights as to where NIRP is having an impact.  The study looked at the cumulative change in the above mentioned variables, expressed in annualized standard deviations.

Looking at Denmark, where we have the longest history of life down the rabbit hole, we can see that NIRP did little to boost wages over its first couple of years,  though the impact on house prices was more immediate and substantial.   Banks outperformed over most of the sample period, though a word of caution: only two banks (Danske and Jyske) were in the sample basket, so it may be difficult to distill too many home truths from such a narrow sample.

In Sweden, we observe not-dissimilar trends over a narrower sample period.   House prices have been a notable beneficiary of NIRP, rising an impressive 3 SDs over just three quarters after its imposition.  Of course, given that housing was already en fuego, the desirability of this outcome is debatable at best.  Wages and banks have both done OK since the depo cut into negative territory, though the margin is slender and the sample size tiny.

So really, the results are unsurprisingly inconclusive; at this juncture we just don't have enough data to make sweeping judgements.  That being said, there are a couple of trends that have emerged that are strong enough for us to distill conclusions.   It seems very clear indeed that NIRP has a depressing impact on bank lending margins; at this juncture it is too early to say that it is posing a threat to financial stability, but recent returns do not look good.   At the very least, it suggests that European banks under NIRP should struggle to perform well, like any other business feeling the pressure of substantial margin compression. 

The impact upon housing seems fairly clear; what better place to park large sums of cash that are under threat of implicit confiscation by the monetary authorities?  Again, how well that plays out remains to be seen.  After all, the Fed tried goosing house prices as a way of climbing out of the post-TMT doldrums a dozen-plus years ago, and didn't that end well?

Whatever the merits of NIRP, at the very least we can say that there is a clear negative externality in the compression of bank margins at a time when global credit creation via the lending channel remains sluggish.  Moreover, there is nothing that NIRP can do about global oversupply across a range of markets, other than to further distort the cost of capital so as to potentially incentivize more mis-investments (which could ironically exacerbate the oversupply issue.)

It seems fitting to close with another quotation from Lewis Carroll, this time via the Queen of Hearts:

"My dear, here we must run as fast as we can, just to stay in place.   If you wish to go anywhere you must run twice as fast as that."