The last couple of months have been quite busy for me with a lot of traveling and particularly the Ukrainian crisis has taken a lot of time so I have not had enough time – or energy – to blog. But don’t worry – I hope that I soon will be back to getting more blog posts out every week. After all as long as we have central banks monetary policy failure is a given and I love to blog about it – even when it makes me very angry.

I had really been planing to write something about Thomas Piketty’s book Capital in the Twenty-First Century. I got it in the mail a couple of days ago, but frankly speaking I can’t really concentrate on reading the book – and I must also admit that his many references to Karl Marx does not make the book more enjoyable (to me at least).

So instead this blog post is an attempt to get over my minor writer’s block by writing a bit about two very interesting working papers I came across while sitting in Stockholm airport earlier today.

The first one is a rather brilliant application of econometrics in the study of German political-economic history. The name of paper is “Highway to Hitler” by Nico Voigtländer and Hans‐Joachim Voth.

This is the abstract:

Can infrastructure investment win “hearts and minds”? We analyze a famous case in the early stages of dictatorship – the building of the motorway network in Nazi Germany. The Autobahn was one of the most important projects of the Hitler government. It was intended to reduce unemployment, and was widely used for propaganda purposes. We examine its role in increasing support for the NS regime by analyzing new data on motorway construction and the 1934 plebiscite, which gave Hitler great powers as head of state. Our results suggest that road building was highly effective, reducing opposition to the nascent Nazi regime.

Extremely interesting…read it!

The second paper is on monetary policy – it is a working paper – The macroeconomic effects of monetary policy: a new measure for the United Kingdom – from the Bank of England by James Cloyne and Patrick Hürtgen.

Here is the abstract:

This paper estimates the effects of monetary policy on the UK economy based on a new, extensive real-time forecast data set. Employing the Romer–Romer identification approach we first construct a new measure of monetary policy innovations for the UK economy. We find that a 1 percentage point increase in the policy rate reduces output by up to 0.6% and inflation by up to 1.0 percentage point after two to three years. Our approach resolves the price puzzle for the United Kingdom and we show that forecasts are crucial for this result. Finally, we show that the response of policy after the initial innovation is crucial for interpreting estimates of the effect of monetary policy. We can then reconcile differences across empirical specifications, with the wider vector autoregression literature and between our United Kingdom results and the larger narrative estimates for the United States.

I think that the method Cloyne and Hürtgen use to study the macroeconomic impact of monetary policy shocks is exactly the right method to use. I have always had a lot of sympathy for the so-called narrative method pioneered by Christina and David Romer in 2004. Cloyne and Hürtgen’s paper is inspired by the Romers’ approach.

This is from the summary of the paper:

Identifying the effects of changes in monetary policy requires confronting at least three technical challenges. First, monetary policy instruments, interest rates, and other macroeconomic variables are determined simultaneously as policymakers both respond to macroeconomic fluctuations and intend their decisions to affect the economy. Second, policymakers are likely to react to expected future economic conditions as well as current and past information. Third, policymakers base their decisions on “real-time” data (that available at the time), not the ex-post (revised) data often used in empirical studies.

A major advantage of the Romer and Romer approach is that we can directly tackle all three of these empirical challenges. First, we need to disentangle cyclical movements in short-term market interest rates from policymakers’ intended changes in the policy target rate. A particular advantage of studying the United Kingdom is that the Bank of England’s policy rate, Bank Rate, is the intended policy target rate. We therefore do not need to construct the implied policy target  rate from central bank minutes as in Romer and Romer did. As a second step, the target rate series is purged of discretionary policy changes that were responding to information about changes in the macroeconomy. This may include real-time data and forecasts that determine the policy reaction to anticipated economic conditions.

Yes, (market) monetarists might say that we should not focus (exclusively) on interest rates, but the authors are nonetheless right to do it in the case of Bank of England as the interest rate has been the key policy “instrument” (actually an intermediate target) for the BoE in the period studied in the paper.

I think we can draw two very clear conclusions from the paper. 1) We should think of monetary policy shocks are deviations from the central banks’ announced rule/reaction function and we cannot say monetary policy is easy is interest rates are low. Monetary policy is only easy if the key policy rate is lower than what it should be according to the policy rule. 2) Monetary policy is extremely potent.

I should, however, also say I missed two things in the paper. 1) The paper only looks at the period until 2007. It would have been extremely interesting to see what happened in 2008. My expectation would be that the method used in the paper would reveal that the British economy was hit by a major negative monetary policy shock in 2008. The BoE failed. 2) I would have loved to see the method applied to “unconventional” monetary policy (I hate that term!) – has BoE continued a clearly defined rule after 2007? I think not…

Enjoy both of these rather brilliant working papers…