What has been driving the real exchange rate?

For anyone that has been looking at posts over here, or carefully listening to Reserve Bank speeches, the topic of the real exchange rate is an important one for understanding the New Zealand economy.  Many of the “concerns” or “issues” being raised at present are really just a function of some view of the real exchange rate.

Via the RBNZ we have a graph of the real exchange rate (RER) here:

Now this drives the question, what has caused the change in the real exchange rate – what shocks have we experienced that have pushed it up, and what proportion of the increase was due to these shocks.  Chris McDonald at the Reserve Bank decided to have a go at answering that question.  With so many factors driving the dollar, “causation” is hard to appropriately appropriate between causes – and so his primary focus is on the correlations and their magnitude, albeit within a framework that will help to show what the more important drivers are.  So what is his conclusion:

• International factors relevant to New Zealand explain more (60 percent) of the exchange rate variance over our sample than idiosyncratic and domestic factors.
• The most important international factor is likely to be export commodity prices, though our empirical analysis is not conclusive. For instance, high commodity prices can explain why the exchange rate is at current high levels. But, high commodity prices may be partly a result of current low foreign interest rates.
• The best domestic indicator for the exchange rate is house price inflation. While this indicator also reflects international factors, its movements over and above the impact of these appears to capture some key domestic information for the exchange rate.

Now this doesn’t tell us anything about the key issue of the New Zealand dollar being “persistently overvalued” or not.  But it does indicate the commodity prices have been a major driver of the increases we have seen.  On top of that another interesting point was raised:

The RER response to the other domestic shocks suggests some of them may not be well identified. Notably, an unexplained fall in the 90-day interest rate and an unexplained fall in the output gap both have little impact on the RER. Practically, we expect these shocks to cause quite large movements in the RER. However, once we allow for the correlation of these variables with the international and New Zealand real house price inflation variables, these shocks (despite being not so well identified) have a relatively small impact on the results.

So within this decomposition, the impact of a monetary policy shock (change in 90day bill rate) or exogenous change in AD (fall in output gap) are poorly identified – and seem to have little impact on the RER.  The author believes thsi result doesn’t pass the smell test, which is fair enough – after all the author has the best knowledge of what their empirical model is saying (especially since no empirical results are included).  However, if we were to take it at face value it would suggest that the RBNZ’s ability to actually change the RER with monetary, even in the relatively short term, is limited.