news article background image
2018-01-25Macro `n Cheese

Dependence day

In this week’s blog Roger takes a look at some of the arguments for central banks staying independent. Not from politicians, though, but from each other. Alas, for the smaller ones, it might not be as easy as just having a flexible currency.

Over the years there have been many discussions on the best stabilization policy for small open economies. A fixed exchange rate was long held as the best choice, but recurring currency and financial crisis invalidated that policy description. Currently, the conventional – trilemma – view is that under free capital flows a country can only pursue an independent monetary policy if it allows for flexible exchange rates.

Lately, after the global financial crisis, that view has been challenged by, inter alia, Hélène Rey (2015) who argues that independent monetary policy is only possible if capital flows are managed in some way. However, that will be no easy feat. Which is why monetary policy spill-over from center to periphery will surely continue. But, how big is the spill-over and if it is large, what would that imply for stabilization policies in small open economies? Is there, exempli gratia, even any reason to remain outside the Euro Area for countries such as Switzerland and Sweden?

These questions become even more pertinent now that the European Central Bank (ECB) is widely expected to embark on a path towards monetary policy normalization. To what extent could the Swiss National Bank (SNB) and the Riksbank be expected to follow suit?

ECB effects on Euro Area developments

Ideally, we would like to perform a (block-restricted) structural VAR analysis like Canova (2005) but in Macrobond, at least for now1, we structure our VAR-models through careful ordering. Here I use not only Canova, but also Mojon & Peersman (2001) and Kucharčuková et al (2014) and others for inspiration. I try out a number of monetary policy aggregates, including Wu-Xia’s (2015) estimate of the Main Refinancing Operations Rate (MRO), which also take into account the effects of the different unconventional monetary policy actions. However, given that I would like to have a similar measure for the ECB as well as the SNB and the Riksbank I have settled with using the different interbank rates when comparing across countries. For activity I have chosen industrial production2 instead of GDP as it is available on a monthly basis, which increases the number of observations and for inflation I have chosen the target inflation variables for ECB (HICP), SNB (CPI) and Riksbank (CPIF)3. Furthermore, the information criteria suggest that one or two lags should suffice. We use two lags throughout our analysis. To begin, let us look at the impulse-response functions (IRFs) for the Euro Area.

1Keep those support tickets coming!

2For Switzerland I, in the end, opted to use the Business trend indicator that has a 0.9 correlation with industrial production on a quarterly basis, but is available also on a monthly basis.

3To control for international demand and prices, I have also added a global industrial production index and a commodity price index, used throughout. This also dampens the “price puzzle” effects discussed below.

Macrobond Document Macrobond Document
Macrobond Document Macrobond Document
Macrobond Document Macrobond Document
Note: The Impulse-Response Functions are cumulated and timeless but are in Macrobond, for simplicity, expressed as time series and forecasts. Here I have differentiated the series for a, hopefully, more intuitive interpretation.
Charts 1a-f: Euroland impact of ECB-policies defined in terms of Euribor (left) and Wu-Xia MRO-rate (right)

The results show that a monetary tightening results in a quite rapid drop in both industrial production and inflation and the effects peak after 15 and 20 months respectively. This is in line with many other studies on the impact of ECB monetary policies. As can be noted, our results demonstrate the oft-occurring “price puzzle”, with prices rising for some time after the innovation to monetary policy (a hike). This is also a visible effect, albeit very short-lived, for industrial production4. As in previous studies, adding further controls weakens the price puzzle. The only way to completely cure the price puzzle in our calculations is, however, to shorten lag length to 1 (higher lags increase the price puzzle). Using the Wu-Xia shadow MRO-rate (charts in the right column) does not alter this view in any dramatic way, but we do notice that peak effects come somewhat later and that the price puzzle seems to diminish, which is probably an effect of the Wu-Xia estimate containing more information about the future inflation path (since it takes UMP into consideration) than the EURIBOR.

Is ECB-policy contagious?

I now turn to the impact of ECB monetary policy on Switzerland and Sweden, and contrast it to the effects of domestic monetary policy. Ideally, the effects of ECB-policies would in these calculations be unaltered in comparison to the estimates in chart 1 (the reason a block restricted structural VAR is preferable). A swift comparison will show that this is not the case, despite my careful structuring. However, the signs are consistent with previous calculations and the change in coefficients and their subsequent effects on our Switzerland and Sweden models might even cancel each other out – we simply cannot say without more advanced IRF-functionality in Macrobond. Other than that, the most apparent difference when adding Switzerland and Sweden respectively is that the Euro Area peak effects come a few months earlier than in the pure Euro Area model above, which is more or less the same effect that controlling for international demand gives rise to (suggesting that, e.g., there could be an extra-Euro Area component to Swiss and Swedish demand as well). With that, let’s move on to the main purpose, to see if and how independent the SNB and the Riksbank really are.

4Using GDP instead of IPI to a large extent solves this puzzle, which indicates that we have not fully controlled for domestic, service sector, effects.

Switzerland

Macrobond Document Macrobond Document
Macrobond Document Macrobond Document
Macrobond Document Macrobond Document
Macrobond Document Macrobond Document
Charts 2a-h

A monetary tightening by ECB leads to an immediate drop in Swiss industrial production and, within a few months, inflation – much the same way that a SNB-hike would. The transmission is rather rapid with the peak effect occurring after 12 months, and 22 months, respectively. Compared to a hike by the SNB, the effects nonetheless seem somewhat delayed, more so when looking at inflation, but the price puzzle is apparent no matter if ECB or SNB hike rates.

The relative delay in effects from an ECB hike (compared to a SNB ditto) is probably the result of the subsequent hike by the SNB. In addition, it is likely that the franc-depreciation following an ECB-hike boosts up-stream import prices and, later, also consumer prices (inflation). This should provide ample explanation to the relative delay in effects from an ECB- vs a SNB-hike. The fact that a SNB-hike also results in an immediate weakening of the franc should not detract from that conclusion as the long-term effect of a SNB-hike is a stronger franc and the initial lag could, arguably, be an issue of to what extent an SNB-hike suffices to balance a previous ECB-action (not (sufficiently) controlled for in our calculations). Possibly, the frequent use of UMP in the post-GFC period can also be related to these results, or it could simply be due to any other information that we have not managed to control for.

Sweden

Macrobond Document Macrobond Document
Macrobond Document Macrobond Document
Macrobond Document Macrobond Document
Macrobond Document Macrobond Document
Charts 3a-h

The take-away when looking at the IRFs for Sweden is more or less the same as for Switzerland. Admittedly, the IRFs are not as well behaved as for Switzerland but after a few months the effect on industrial production and inflation from an ECB hike are clearly negative and the depressing effect on inflation is even swifter than for Switzerland. Still, compared to a Riksbank-hike, the effects are slightly delayed. (Though less so than what the calculations for Switzerland suggested.)

In line with what could be expected given the very strong focus on the Krona and use of UMPs over the past years, we can see that a hike from ECB would only yield a very small initial positive response in domestic rates. Over most of the response-period it also seems as if the Riksbank’s lukewarm reaction drives interest rate spreads wider and the SEK thus remains long-term lower (helping, perhaps, to explain the quite positive initial response of Swedish industrial production to an ECB-hike?).

If you can’t beat them?

The conclusions from our simple exercise is that ECB-policies do spill over onto both Switzerland and Sweden. The effects from ECB-policies are similar in that they propagate domestic monetary policy actions. Deviations in transmission are probably the result of subsequent reactions from domestic monetary policy makers, but the medium- to long-term effects from ECB-policies are analogous for both Switzerland and Sweden. Hence, the ability to indefinitely deviate from ECB-policies and conduct a truly independent monetary policy seems limited, as work by Hélène Rey (2015) and others have suggested.

However, even though it might not be possible in long-term to conduct a fully independent monetary policy, some leeway exists, which both Switzerland and Sweden have demonstrated in the post-GFC period. This flexibility has, admittedly, become strained as conventional policy has become exhausted begging the question if there are other, structural, forces at work that monetary policy is not fit to address, regardless of where it is conducted. In that case, neither out-hiking nor doubling down on UMP is the best way forward even if it seems warranted by, e.g., continued low-flation.

As stated, many academics are again suggesting capital controls could be utilized - in particular such that mitigate negative effects from the global financial cycle, in itself a result of the monetary policy of main central banks. The remedy, it is suggested, is this: Targeted capital controls; influencing global monetary policy makers, and; increased cyclical and structural buffers.

To me, remembering policy recommendations for small open economies from a time passed, it seems instead that we will soon have come full circle. Might now be a good time to bring up Bancor again?


Get these posts delivered to your inbox

If you like what you read, why not subscribe to have these analyses delivered to your inbox every Friday?

Sign up


Disclaimer

We don’t usually have views and opinions about economic and financial states of affairs, (not ones that we express publicly as a company, anyway). We do believe, however, that people can and do appreciate a variety of perspectives. What you’ve just read is the perspective of our resident chief economist. While we think he’s very smart, Macrobond Financial does not expressly endorse the views he presents here. And, as the old adage goes, you shouldn’t believe everything you read (not without finding the data, performing a few analyses and presenting it in a nice chart). We want to make it clear that we are not offering this information as investment advice. That being said, if you have the application you can easily check everything that’s mentioned here, and decide for yourself. If you don’t have the application, now you have a great reason to get it.