Markets are telling us where NGDP growth is heading

I am still in Provo Utah and even though I have had a busy time I have watch a bit of Bloomberg TV and CNBC over the last couple of days (to fight my jet lag). I have noticed some very puzzling comments from commentators. There have been one special theme and that has come up again and again over the last couple of days among the commentators on US financial TV and that is that “yeah, monetary easing might be positive for the markets, but it is not have any impact on the real economy”. This is a story about disconnect between the economy and the markets.

I find that perception very odd, but it seems like a lot of commentators simply are not mentally able to accept that monetary policy is highly effective. The story goes that when the Federal Reserve and the ECB moves towards monetary easing then it might do the markets good, but “real people” will not be helped. I find it unbelievable that well-educated economists would make such claims.

Markets are forward-looking and market pricing is the best tool we have for forecasting the future. When stock prices are rising, bond yields are rising, the dollar is weakening and commodity prices are going up then it is a very good indication that monetary conditions are getting easier and easier monetary conditions mean higher nominal GDP growth (remember MV=NGP!) and with sticky prices and excess capacity that most likely also mean higher real GDP growth. That has always been that case and that is also the case now. There is no disconnect between the markets and the economy, but there is a disconnect between what many commentators would like to see (that monetary policy is not working) and the reality.

To try to illustrate the connection between the markets and NGDP I have constructed a very simple index to track market expectations of future NGDP. I have only used two market indicators – a dollar index and the S&P500. I am constructed an index based on these two indicators – I have looked year-year percentage changes in both indices. I have standardized the indices and deducted them from each other – remember higher S&P500 means higher NGDP, but a stronger dollar (a higher USD index) means lower NGDP. I call this index the NGDP Market Indicator. The indicator has been standardized so it has the same average and standard deviation as NGDP growth since 1990.

As the graph below shows this simple indicator for future NGDP growth has done a fairly good job in forecasting NGDP since 1990. (You can see the background data for the indicator here).

During the 1990s the indicator indicates a fairly stable growth rate of NGDP and that is in fact what we had. In 1999 the indicator started to send a pretty clear signal that NGDP growth was going to slow – and that is exactly what we got. The indicator also clearly captures the shock in 2008 and the recovery in 2009-10.

It is obvious that this indicator is not perfect, but the indicator nonetheless clearly illustrates that there in general is no disconnect between the markets and the economy – when stock prices are rising and the dollar is weakening at the same time then it would normally be indicating that NGDP growth will be accelerating in the coming quarters. Having that in mind it is of course worrying that the indicator in the last couple of months has been indicating a relative sharp slowdown in NGDP growth, which of course provides some justification for the Fed’s recent action.

I must stress that I have constructed the NGDP market indicator for illustrative purposes, but I am also convinced that if commodity prices and bond yields and maybe market inflation expectations were included in the indicator and the weighing of the different sub-indicators was based on proper econometric methods (rather than a simple unweighted index) then it would be possible to construct an indicator that would be able to forecast NGDP growth 1-4 quarters ahead very well.

So again – there is no disconnect between the markets and the economy. Rather market prices are very good indicators of monetary policy “easiness” and therefore of future NGDP. In fact there is probably no better indicator for the monetary policy stance than market prices and the Federal Reserve and other central banks should utilize market prices much more in assessing the impact of monetary policy on the economy than it presently the case. An obvious possibility is also to use a future NGDP to guide monetary policy as suggested by Scott Sumner.

Related posts:

Understanding financial markets with MV=PY – a look at the bond market
Don’t forget the ”Market” in Market Monetarism
Central banks should set up prediction markets
Market Monetarist Methodology – Markets rather than econometric testing
Brad, the market will tell you when monetary policy is easy
Keleher’s Market Monetarism

Leave a comment


  1. jpirving

     /  September 29, 2012

    Nice index Lars, simple and gets the job done. I’ve tried on and off for a while to run an NGDP forecasting model off only market prices (a VAR) and simulate a ‘historical’ forecast by hiding part of the data set from the parameter estimation. It has so far not given terribly credible historical NGDP expectations, but your index seems pretty good. It would be nice to somehow hook this up to live market prices and have it run in real time off market tick data on a day to day basis. Extracting a synthetic NGDP expectation, from market prices would be highly valuable for research and testing the market monetarist model, me thinks.

  2. Saturos

     /  September 29, 2012

    Lars I like your indicator but you are being unnecessarily indirect. We care about RGDP even more than NGDP, and equity prices signal RGDP directly. What use are our financial commentators on the news if they can’t even explain that stock prices are society’s best estimate of its future real income?

  3. Lars,
    I have been completely fascinated by your spreadsheet since I found this post. This combination of techniques — subtract the series average… divide by the series standard deviation… multiply by a different series standard deviation… and add that other series average — is there a name for this set of techniques? Is it commonly done, or is it something that you devised? (I have been calling it a “Christensen Fit” calculation.)

    After several false starts, I have evaluated the arithmetic in the spreadsheet you link to. There are some things in the spreadsheet that I do not understand the reason for, in particular: subtracting 1.5 after adding the NGDP series average. Perhaps my evaluation would be of interest to you:

  1. Ben maybe you should try “policy futures”? « The Market Monetarist
  2. Adding market expectations to our models « Economic Sophisms
  3. Bringing NGDP expectations into the picture | Historinhas
  4. Gabe Newell on prediction markets – NGDP level targeting and Lindsay Lohan « The Market Monetarist
  5. Unfocused vacation musings on money – part 2 | The Market Monetarist
  6. Justin Irving’s real-time US NGDP indicator | The Market Monetarist
  7. “Markets are telling where NGDP growth is heading” – an update | Historinhas
  8. Prediction market: Fed on track to hit 4% NGDP growth in 2015 | The Market Monetarist

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: