ECB QE and external flows
A nod’s as good as a wink to the markets
Mr Draghi’s Sintra speech was more hawkish than markets had expected following a fairly dovish June press conference. Despite some backtracking, the ECB failed to get the toothpaste back in the tube, and so the euro has rallied since then. We use the ECB’s monetary statistics to try to get perspective on this.
Fed’s Brainard on balance sheets and FX
Further light on the relationship between various monetary policy tools and the exchange rate was cast in a recent speech by Fed Governor Lael Brainard. Governor Brainard is probably one of the FOMC members who takes most note of the relationships between the US and global economies. She examined the potential spill overs from balance sheet action with the consequences of raising short rates. In brief, Brainard argues that since exchange rates appear to be more sensitive to short rates than long ones, a given quantum of domestic tightening will affect the exchange rate by less if it is accomplished through balance sheet run-off as opposed to raising short rates. This sensitivity to short rates has been a key focus for BNPP’s FX forecasts for a long time. It may also help to understand why the ECB internal debate on sequencing short rates and balance sheet actions went the way it did in the spring.
Balance sheets affect FX less than short rates
We have believed for some time that one of the possible reasons the Fed has been so much in favour of reducing its balance sheet is precisely to avoid a too-strong dollar, which we have argued, could put the Fed at odds with the White House if the greenback were to hamper the revival of US manufacturing that President Trump clearly wants to achieve. Brainard’s speech does not tackle this issue directly, but the analysis fits with our conclusions.
Is avoiding FX moves reason for some co-ordination?
Governor Brainard also makes the case that simultaneous balance sheet action by two countries could leave exchange rates more or less unaffected. This could help to reduce the risks of too “rich” asset prices, while at the same time avoiding sharp exchange rate shifts that could impede central bank achievement of objectives.
We examine QE and external flows
Governor Brainard looks at the issue through the lens of interest rates. We take a different perspective here by examining how ECB QE has affected the composition of the balance sheet of the banking system. The banks’ balance sheet adds up, so we can examine the impact of increased credit to the government (in recent years, principally QE) on other items in the balance sheet – monetary growth, for example, or credit to the private sector.
In the current instance, we explore the relationship between credit to the government (QE) and the banking system’s net foreign assets in the euro area.
How will QE affect net foreign assets?
Net foreign assets are the foreign assets of the banking system less its foreign liabilities. When a central bank buys government bonds, at least some will be bought from foreigners. In the first instance the foreigner will receive a bank deposit (a foreign liability of the banking system in the country where QE is taking place) and so net foreign assets will decline (initially, one-for-one with the quantity of bonds purchased from foreigners). Unless there are second-round effects, there will be no change in the money stock. When the central bank buys a bond from a domestic resident, the resident’s bank deposits increase, again one-for-one in the first instance. This will raise the money stock, but will not affect net foreign asset (NFA). Second-round effects can change this – if the domestic resident were to buy bonds from foreign hands with her/his proceeds, the NFA position of the banking system would fall and the money stock would return to its pre-QE level.
Credit to government has driven M3 growth…
Turning to the ECB, in addition to the signalling effect of Mr Draghi’s speech on expectations of monetary policy, Chart 1 can help to explain why ending ECB QE could have an important effect on the currency. There are multiple other factors, such as overseas developments that affect our forecasts, but we are focusing only on this one for now. The chart shows the counterparts of monetary growth in the euro area. M3 represents the monetary liabilities of the banking system. The counterparts approach starts from the fact that the banking system’s balance sheet balances. Thus, movements in money can be explained by shifts in the other items of the balance sheet.
…and led to very negative external counterparts
What the chart show is that monetary growth has been pretty steady over the last couple of years. The major counterpart to this has been credit to government (QE in other words). Currently, roughly 3½pp of M3’s 5% growth rate is being driven by credit to government. It is very noticeable that as credit to government stepped up in 2015, the external counterparts to money became more negative. Think of these as being outflows from the euro area. In short, QE caused external outflows and weakened the euro.
Ending QE will slow M3 and require stronger EUR
When QE is ended, the contribution that credit to government will make to monetary growth will very probably fall, though not necessarily to zero since commercial banks still may buy government debt. Since credit to the government has accounted for such a large amount of money’s 5% growth over the last year, monetary growth would slow, possibly significantly, all other things equal. But other things are unlikely to be equal: the external counterparts will likely cease being so negative and could turn positive.
Nearly half QE bonds bought from foreigners
One way of reviewing this is to say that a good proportion of ECB QE is financing outflows, and when this stops, the EUR will rise. Since the ECB reckons about 45% of its purchases made in the two years through March came from foreign residents (ECB Bulletin June 2017), it is easy to see why there could be a big exchange-rate effect, even before we take into account the possible purchases of overseas assets by domestic residents following QE.
M3 growth could slow sharply
We viewed this phenomenon through a different prism, leading to the same conclusion. When ECB QE stops, the sharp slowdown in M3 we would expect through the counterparts analysis will most likely lead to an insufficient growth in money supply to meet the demands of the growing economy. This will put upward pressure on the exchange rate, which would reduce the price level relative to what it would otherwise be and bring the real money supply back into alignment with money demand, also slowing growth along the way.
QE reduction will strengthen EUR
Whichever way you look at it, it is difficult to avoid the conclusion that reduced ECB QE will – all other things being equal – be a force for a stronger euro. Other things are not equal, however. In particular, we know that the Fed is set on reducing its balance sheet beginning later this year. Other factors also affect BNPP’s currency forecasts, but we see QE as having been a major driver over the period since 2015, and it may well be again going forward. There is an open question as to whether QE and negative rates together have a bigger FX impact than QE in the presence of positive rates.
Chart 1: Counterparts to Eurozone money growth (% y/y)
Chart 2: External counterparts to money and EURUSD
BNP Paribas research reports as well as market updates from major investment banks are available to our regular subscribers.