Kaleidic Economics has been flirting with prediction markets since our launch. I have also been intruiged by Scott Sumner's proposal for an NGDP futures market (see this proposal for the Adam Smith Institute), and the impact this would have on the discretion of central bankers. As I said in my brief review for Money Marketing:
Targeting the level removes a large amount of discretion from the Bank of England. Monetary policy stops being an arbitrary policy lever and provides a macroeconomic foundation upon which economic activity can build.
Some time ago I set up a prediction market for NGDP using Inkling Markets. Here are the details:
The main problem with this is the difficulty of finding NGDP figures for the UK. Britmouse provides this useful guide to UK GDP but my suggestion is to start off with Table A1 of the Quarterly National Accounts (published by the Office for National Statistics). It would be handy if there was a stable link for the most recent release, but the best I can find is going to "All releases of Quarterly National Accounts". You can then click on the most recent, and eventually "select series from this dataset". Table A1 has what you need.
I think that it's easiest to think of NGDP in terms of quarterly growth rates that are compared to the previous year. This was what I had in mind when I posed the question, and here are the components (the red and blue lines should sum to the orange one):
NGDP = Real GDP + GDP Deflator
IHYO = IHYR + IHYU
The chart below shows the composition of NGDP over the last decade.
You can also do this for annual figures (YoY change)
IHYM = IHYP + IHYS
Or quarterly figures (QoQ change)
IHYN = IHYQ + IHYT
As Lars Christensen rightly pointed out it is the level of NGDP rather than a particular growth rate that "market monetarists" care about. One of the downsides of NGDP targetting is that people tend to think in terms of growth rates rather than levels. But that can change. Therefore it would probably be better to focus on the level of NGDP:
The prediction market question should probably be in relation to this figure, therefore Kaleidic Economics will endeavour to update it on a quarterly basis as part of our data. The series code is YBEU. I'm open to suggestions for what that question should be - for example an annual figure on an annual basis? Or a quarterly figure on a quarterly basis? For now I've done the following trial:
Finally, for the absolute amount of NGDP:
- Table C1: Quarterly National Accounts - Gross domestic product: expenditure at current market prices = YBHA
Other Kaleidic resources:
- "The perils of NGDP targeting", Quarterly Report No. 2, December 2011