It used to be a bane for short-term rate models to have negative interest rates. Cash has alternatives besides holding debts with a fixed income and naturally, creator of the closed form analytical formula to tight pricing options with its sensitivity parameters to market variables and its underlying, Fisher Black, used the max(Shadow rate,0) to create a Zero-Level Bound on policy rates, on which banks in the country and countries who pegged against them benchmark. Nevertheless, desperate times call for unconventional policies and BOJ, SNB and ECB started negative policy rates, in which they assumed there is a “certain threshold” of “storage cost of cash in banks instead of paper form” that people are willing to pay. Its to counteract zan aggregate narrowing output of domestic savings, investment and consumption choices.
That said, the ZLB floor is obsolete and interest into Shadow rate was regained. Shadow rate is another term for the underlying short-term rate modeled by tree rate processes. Very similar to BS model on option pricing, just that the only difference is, tree process is not on underlying asset price but on short-term rates. Main differences in nature of the tree process are 1) that debts are redeem at par hence a pull to par value at maturity 2) that payoff or discount factors (flip side of expected returns assuming debt held till maturity) is derived from par swap rates 3) that while pricing equity options assume random volatility per tree step, rates volatility and convexity (think options gamma) decline as debt nears redemption. Drift factor, or expected return, in short-term rates tree process refers to the cumulative-per-step risk premium on top of YTM (aka starting node of tree) and volatility diffusion (dw) should fade out in a exp(T-t) process as it nears maturity. This naturally gives the yield curve a upward slope at the short end till belly of the curve (note: has been pushed forward given long period of ZLB policy rates in most G10 countries) and a mildly downwards slope on the long end given that convexity arising from volatility overwhelms the cumulative positive risk premium demanded from investors.
But the fun thing in economics, its more a art than a science. As George Soros put it, “in the cognitive function reality is given; in the participating function, the participant’s understanding is the constant”. Economist painfully recognised that the economy is made up of “thinking” participants that will not stand on a railway track and not move when the train is approaching fast towards them.
Investors react to central bankers actions, and so the other way too. A bump up on economic growth forecast can see equities aggregate market-cap weighted ROE being push up corresponding, even before earnings season. Unless its really unpredictable and the expected returns for holding the risk in not worthwhile, none would take a wait-and-see approach. Thus, an impulse response or a monetary policy transmission analysis was done by Jing Cynthia Wu and Fan Dora Xia.
Summary of their paper
These forward rates are constructed with end-of-month Nelson-Siegel-Svensson yield curve parameters from the Gurkaynak, Sack, and Wright (2006) dataset. The full details of the Wu and Xia model are described in their accompanying working paper. In short, the shadow rate is assumed to be a linear function of three latent variables called factors, which follow a VAR(1) process. The latent factors and the shadow rate are estimated with the extended Kalman filter.
Shown below, a -25bp shock should boost macroeconomic data to the upside. Under Fisher’s ZLB rule, things would have worked perfectly fine in 1960-2013. However, post GFC (2009-2013) things changed and the response were unfelt such that conventional policies were deemed ineffective, prompting to a new policy regime. FAVAR (Factor-Augumented Vector Auto Regression) are widely used to trace out the transmission effectiveness of monetary policy innovations on the economy. To makes things worse, they were comparing a YoY 13 months lag FAVAR to MoM 1 month lag FAVAR under the new policy rate i.e. at ZLB. Employees, new house owners and producers were the best immediate beneficiaries, whereas the aggregate economy as an entity, had persistent lackluster IP growth and were under a long period of deflationary pressure, an outcome of families cutting back on purchases and a failure of producers to recognise the huge change in saving-spending habits and demographic shift and cut down investories and production, resulting in credit and output imbalances.
As we should have known, Shadow rate, the unfloored modeled rate, is different from the observed policy rate, FFR. Realised polcy rate did in fact plunged throughout 2010 till 2014 so that the other macroeconomic variables can see improvements.
In fact, till end 2014 only did the shadow rate pick up and surpass +25bp. When shadow rate is below 0, central bank will hold on to rates at ZLB. As expected, when shadow rate broke above +25bp upper band of corridor (i.e. target, not setting policy rate) Fed started its hike of FFR. This tightening of credit supply also was the reason equities (assets) suffered in 2015.
Now that the Fed has taken the lead of setting a hawish tone among central bankers, we saw BOC as well as ECB carrying hawish tone. Why? Not doing so might give investors the expectations that their economy is still weak, their currencies might be devalued that improve their domestic export competitiveness but hamper the other side of BOP: a capital outflow to countries that provides higher yields.
That said, the chart below shows that ECB should not be hiking rates in the near future, given that shadow rate is nowhere above the ZLB. While PMI have been improving and more are accepting an upbeat domestic growth forecast, it seems that this year could be similar to what US were facing in 2014. If ECB can see a similar aggressive tightening in credit supply as seen in US 2014 huge rise in her Shadow rate, 2018 could well be the time we see ECB returning from negative rates to more conventional policies, keeping its deposit facility rate above ZLB.
Source: ECB, Wu-Xia Shadow Rate
Given current market pricing of its policy rate, ECB is well before Fed in in terms of unwinding its balance sheet. It would take ECB untill Dec 2018 to see it above -0.20bp, till then, we can monitor Wu-Xia shadow rate on ECB deposit facility rate to see if its picking up.
Yes, market turned slightly dovish when EURUSD broke above 1.15. This is the level ECB usually verbally intervene and talk down hawkish expectations.