I seem to be constantly quoting from the FT, but I can’t help it – today’s article on Alphaville was a very good one in my opinion. Here Peter Doyle writes about the lessons we can learn from the oldest central bank, the Riksbank in Sweden, which cut its key interest rates to 0.00% last week.
The Riksbank depends on IMF global forecasts to determine its policies, and the lesson here according to Mr Doyle is to be more skeptical, acknowledging that for such forecasts, there is a natural upward bias.
In addition, a particularly interesting tit-bit:
But there is still a lesson to draw from Sweden. Until 2004, it routinely conducted a fantastically detailed and comprehensive household survey, including assets and liabilities, which allowed supervisors to run all sorts of macro, employment, exchange and interest rate, and integrated sector-specific scenarios on their individual and collective banks’ books, tracing effects down to all individual (as opposed to representative) households. This went way beyond the detail, severity, and crude assumptions implicit in the standard “stress testing” typified by the recent Euro Area exercises.
He explained in the comments section that this practice had been discontinued because “There were concerns amongst the wealthy that the data would be used to institute a wealth tax, more recently compounded by post-Snowden concerns with government possession of such data”.
My mother worked as a demographer, so I grew up observing her conducting surveys after surveys. Data is crucial to policy making because otherwise, policy makers would be ‘flying blind’ while determining which direction they should go. I fully agree with him when he writes that “there is critical need for surveys as detailed as those in Sweden to 2004, so that fragility evaluations do not rely on dubious indicators such as house prices, debt to income ratios, and so-called “stress” tests”.
The article closes pondering how forward guidance should be designed, a topic which I have tweeted and blogged about countless times:
Need for central bank guidance has to be weighed against need to allow flexible responses to surprises. But, as Sweden illustrates, that does not mean that “anything goes” on forward guidance. If the bias in major sister central banks and global ministries of finance is to “jump the gun” (to tighten too early/loosen too late) and global forecasters are biased in those authorities’ favour on such judgements, then the structure of forward guidance has to take those international biases into account.
In particular, for any individual central bank, those global biases shift the balance towards erring on the side of accommodating potential inflationary pressures. So guidance might be set to tighten policy only when actual domestic inflation is rising, not when some possible leading indicator of it, such as global forecasts or unemployment or productivity, apparently forewarns. This is what Sweden has now committed to do, having overlooked these international biases earlier.
We shall see whether this will be a wise thing to do in the coming years, the insight being that models should correct for such biases. Having been through an unsuccessful half-tightening cycle Sweden has now signaled very strongly via setting the rate to exactly zero that it has no intention of trying again in the near term. There is hope that other central banks may take away similar lessons: deflation carries far more danger in highly leveraged economies than a spike of inflation that should be easier to control once it actually happens.
With that thought, I will include this story about being rational and following ‘guidance’:
The theater was crowded with guests and there was only one usher. There had only ever been one usher. The crowd was a familiar crowd, and equally familiar with this usher, and knew how incompetent he was, always misdirecting them to the wrong seats.
A fire broke out, the alarm went off and the electricity went out. It was total darkness. The crowd murmured in panic. Suddenly, a tiny light came on, it was the usher. “Never fear, ladies and gentlemen, I have a light!” He had with him one teeny, tiny torchlight that he used for directing. “Follow me,I know the way!”
The short relief the crowd felt was immediately followed by, “should we follow him?” doubts, knowing as they did how incompetent the usher was. For all they knew, the usher might direct them further into the building and straight into the fire. But what else should they do? The usher was the only one who had a light in that cavernous place and he did claim he knew the way. A few began to follow him, and pretty soon everyone followed as a crowd.
From the usher’s point of view, if he walked too fast the crowd might lose him or he might lose part of the crowd. However, if he walked too slowly, the fire might spread and there would not be enough time to exit.
Was it logical to expect that the usher knew the way having prior experience of his incompetence? Was following him the rational thing to do? With limited information and choices, yes, it was rational. Was it wise, who knew?