Not all that much has happened with the USD/EUR exchange rate in recent years. Yes, there have been some sizeable swings from time to time, but on average the movements have been pretty muted. Actually, over the last 12 months there has hardly been any movement at all. The reason for this is that two of the most important factors for the USD/EUR exchange rate have been sending out conflicting signals. However, I will try to explain why this could change in the foreseeable future.
Central Bank Balance Sheets and Short-term Interest Rates
Which two factors am I talking about? It is probably no surprise that these are related to central banks and their policies. Since the outbreak of the financial crisis a considerable part of the USD/EUR exchange rate movements can be explained by
– changes in the size of the Fed’s balance sheet relative to the size of the balance sheet of the ECB, and
– changes in the short-term interest rate differential between the US and Europe.
Why are these factors so important for the USD/EUR exchange rate? Let’s start with the relative size of the balance sheet. An easy way of interpreting this factor is by looking at it as an indication of relative liquidity or relative supply. The central bank that blows up its balance sheet the most, hence prints the most money, supplies more of its own currency, pushing the value of that currency down.
This relationship is shown in the graph below. The black line shows the development of the relative size of the balance sheet. If the Fed prints more money than the ECB, it goes up and vice versa. Therefore, the black line shows the combined effect over time of the three rounds of Quantitative Easing by the Fed and the LTROs and peripheral bond buying of the ECB on the balance sheets. The blue line shows the USD/EUR exchange rate. It’s not bold to state that, at least most of the time, the two lines move in the same direction. When the Fed was printing more money than the ECB, the USD/EUR also went up, meaning the dollar depreciated. When the ECB created more liquidity, the black line went down and the dollar appreciated. Only in the last part of the graph the relationship looks distorted, but I will come back to that in a minute.
First, let’s turn to the second factor, the short-term interest rate differential. In the next graph I have plotted the USD/EUR against the two-year interest differential between the US and Germany. Again it shows a strong relationship. When the short-term interest rate goes up in the US relative to Germany, the dollar becomes more attractive, and the USD/EUR exchange rate goes down accordingly (mind you – the right axis that shows the interest rate differential is inverted).
Thus, shifts in the relative size of the Fed’s balance sheet and the interest rate differential between the US and Europe have been major factors for the changes in the USD/EUR exchange rate. And, equally important, since these factors have oscillated in recent years, the exchange rate did the same.
It is noteworthy that, as can be derived from the two graphs, the two factors have moved in tandem in recent years, working together to push the exchange rate up or down. Unfortunately, this means a little more statistical backing is required here to show that the relative size in balance sheet and rate differential are not two sides of the same coin. So here goes. Since the beginning of 2010 the correlation between weekly changes in the relative size of the balance sheet and the interest rate differential has been 0.19. This is statistically low and suggests that although they have moved in the same direction over longer periods of time, their co-movements from week to week are very limited.
The distinction between the two factors becomes more visible if we run some simple regressions. The results are shown in the table below. In Regression I weekly changes in the USD/EUR exchange rate are regressed on the changes in the relative size of the central bank balance sheet (Fed/ECB). The estimated coefficient is -0.57 and highly significant, which underpins the negative relationship between the two factors. If the Fed prints more money than the ECB, the dollar depreciates. The regression model explains 32% of the movements in the USD/EUR, as measured by the R Square (this statistic shows the explanatory power of the regression model).
In Regression II the weekly changes in USD/EUR are regressed on the second factor, the short-term interest rate differential. The results are similar to that in Regression I. The relationship is negative (-0.11), higher rates in the US translate to an appreciation of the dollar, and highly significant. The model explains a decent 21% of the movements in the USD/EUR.
More importantly, if both factors are used to explain the changes in the exchange rate, both factors remain highly statistically significant. This leads to the conclusion that the relative size of the central bank balance sheet and the interest rate differential are indeed separate factors that help explain movements in the exchange rate. Also, the explanatory power increased to 43%, which is pretty impressive from a statistical point of view.
Change of direction?
Enough, about the statistical stuff. Why do I want to point out that these factors are so important? Well, for instance, the knowledge that the relative balance sheet and the rate differential have a major impact on the USD/EUR exchange rate can be used to explain why the dollar has hardly moved at all in recent months. Instead of working together the two factors have been drifting apart. Contradictory powers have been at work. The Fed continued expanding its balance sheet while the ECB stayed put, but at the same time the interest rate differential held ground, and actually increased a tiny bit.
This brings me to my final and most important point. Chances are increasing that the two factors influencing the USD/EUR exchange rate could be joining forces again in the foreseeable future. While it could take some months for it to actually happen the Fed will start tapering its bond purchases. This means that the steady rise in the balance sheet of the Fed will start to slow. The ECB however, has in recent months maneuvered itself in a position that enables it to increase its balance sheet if necessary. Suggestions of a new round of LTRO or bank programs to help credit growth to small and midsize companies have become more audible. And, since there is no immediate solution for the debt-stricken peripheral European countries, the ECB could at some point in time be forced to start buying government bonds again.
On the other side I expect the interest rate differential to increase. The current US 2-year interest rate is currently 0.35%, still very low because of the ultra-loose policy of the Fed. The difference with the 10-year rate, which is a better reflection of economic circumstances, is over 2%, which is high in historical context. Furthermore, the Fed futures show that from December 2014 and on, there is a probability of at least 50% that the Fed has raised interest rates to 0.50% or more. Finally, GDP growth is estimated to reach 3% in 2015, which I believe is not overly optimistic. The 2-year interest rate will gradually start to reflect all this. (For more on US interest rates please also see my previous article on yields)
In Europe, however, the situation is very different. As mentioned, the ECB actually reinforced its accommodative stance, albeit only verbally by Mr. Draghi. Interest rates will probably not go up, and if they do it will be less than in the US. The Euro Area is expected to grow just over 1% in 2015, and this will be difficult to achieve. Hence, the German 10-year yield is ‘only’ 1.65%, leaving a much smaller difference with the 2-year yield, which currently stands at 0.14%.
To conclude, when looking at the USD/EUR exchange rate it is very helpful to know that there are two major factors at work. The relative balance sheet of the Fed vs. the ECB and the short-term interest rate differential between the US and Europe have had strong explanatory power concerning the movements in the USD/EUR. For some time now these two factors have been cancelling each other out, leaving a very small bandwidth for the USD/EUR to move in. But, somewhere in the foreseeable future these two factors could join forces again, leading to a new uptrend in the dollar.