Today’s Financial Times has a topical article by commentator Martin Sandbu on the supply shock challenge for monetary policy makers.
Many of his points are well made — supply shocks face policy makers with much more difficult choices than demand shocks. They may need to tighten monetary policy and reduce economic growth at a time when economic activity is already weak. They need to think carefully about this. If the supply shock is temporary there is little need to tighten monetary policy and reduce aggregate spending to control inflation. Inflation will rise temporarily but it will then fall. Even if the current supply shocks are long lasting, which those emanating from the war in Ukraine could be, there is nothing in current data to suggest that spending is yet growing unsustainably rapidly. In both the US and the UK nominal spending is in line with pre-pandemic trends. There is no case yet for a fierce tightening of monetary policy to curb inflation. On all this I agree.
But Martin makes what is to me an analytical error, one which rather undermines his case that the Fed, the ECB and the Bank of England should not be raising policy interest rates today. He does not clearly distinguish nominal interest rates and real interest rates. The tightness or looseness of monetary policy depends on real interest rates, not nominal interests rates, and looked at from this perspective monetary policy today is extremely loose and has been loosened further over the course of the past two years. Some increase in nominal rates, of the order of 2%, is needed to bring real interest rates and aggregate demand in line with supply.
To understand this point, take a look at the following chart, showing one measure of real interest rates the gap between nominal interest rates and nominal income growth.
This chart compares the annual nominal earnings growth in the UK (the orange line, all earnings including bonuses) with two measures of interest rates, the Bank of England base rate (the blue line) and the five year UK government gilt rate (the grey line). The difference between earnings growth and interest rates provides an useful measure of the looseness or tightness of monetary policy. When earnings growth is substantially higher than current interest rates, then it is much more worthwhile to borrow for current spending, growth in earnings outstrips the financing costs of borrowing and there are strong incentives for those who can to borrow and spend more.
As an example of the insight offered by this chart, the comparison of earnings growth and interest rates reveals a fairly loose stance of UK monetary policy in 2004-2007 with no tightening of interests rates in response to increasing earnings growth. This was justified by the inflation targeting framework: with additional spending going largely on imported goods and services there was little danger of rapidly rising domestic prices. This though contributed to an unsustainable boom in credit and property prices that exacerbated vulnerability to the financial crisis of 2007-2008.
This chart also reveals an even more substantial loosening of UK monetary policy from 2015 onwards, with the gap between earnings growth and interest rates rising from around 0% in 2013 to 3% in 2018 (earnings growing 3% per year more than market rates of interest) and yet further loosening post pandemic with this gap rising to 5%. This suggests, Sandbu arguments not withstanding, that monetary policy is now extraordinarily loose and there is a need now to raise interest rates to prevent an unsustainable borrowing boom. Agreed, there is little sign of such unsustainable borrowing yet in the data. There is therefore no case for an immediate 5% rise in the Bank of England base rate to restore the tightness of monetary policy back to where it was in 2014. But, such a large rise could be required if very low real interest rates lead to borrowing, inflation and wage growth getting out of control. To prevent this happening, some rise of nominal interest rates is needed, just to restore real rates to more sustainable levels and prevent an unintended loosening of monetary policy that requires then a much more severe subsequent monetary tightening.
How large a rise? This will depend on subsequent economic developments. I would not claim that this chart of real wage growth and interests rates tells the entire story (fiscal developments will be crucial, as will potential slowing consumer expenditure triggered by higher energy and food prices). Still this chart suggests to me that a rise in the Bank of England base rate of the order of 2% is likely to be needed by end year. This is not hawkish, it is simply recognising that current monetary policy is unsustainably loose.
This is not just a UK story. US earnings data, while more substantially distorted by the emergency pandemic response, shows a similar large increase in growth rates compared to nominal interest rates. See nominal wage data on the FRED database (these need transforming into annual percentage changes) . Worldwide, I believe, nominal interest rates have to rise, and better this is done sooner rather than later to avoid the need for subsequent more aggressive policy.