Scott Sumner  

The pound depreciated by 10%. What does that mean?

Is the Limit on Nuclear Liabil... After Brexit - why not unilate...

Commenter Matthew Moore asked the following question:

Any chance of a post on the likely consequences of GBP devaluation? Ejection from the ERM caused a boom. I realise the UK wasn't deliberately overvalued this time, but are there parallels?
In my view the key difference is not the question of "overvaluation" but rather the source of the shock. In 1992, the devaluation was a policy decision, which was (correctly) seen as enabling a more expansionary monetary policy. Hence it had an expansionary impact on NGDP.

This time around the shock was Brexit, which was seen as reducing the real growth prospects for the British economy. One way we know this is that British stocks fell on the news. By the way, in several earlier posts I underestimated the impact of the shock to British equities, by looking at the FTSE100. Commenter Vaidas Urba pointed out that this index includes lots of multinationals with a great deal of foreign earnings. An index of domestic British firms declined much more sharply, comparable to many mainland stock markets. So the expected hit to the UK economy was presumably larger than I originally thought. Mea culpa.

Nonetheless, I continue to believe that the most important effects are global, and that at the global level this was primarily a monetary shock. For instance, the Japanese yen soared in value and the Japanese stock market crashed. Obviously this has nothing to do with the likely long and boring negotiations over tariff rates between the UK and the continent. This is about risk averse investors looking for a safe haven currency, and pushing that currency up so much that deflation may return to Japan.

One of the most complicated issues here is the extent to which the depreciation of the pound reflects the direct impact of the Brexit shock, and the extent to which it reflects actions the Bank of England might take to soften the blow. In my view the bottom line is that NGDP growth in the UK is likely to slow despite the fall in the pound, based on the reaction of their stock and bond markets. However, the BOE is probably expected to do much better than the BOJ, and hence the hit to UK stocks should probably be thought of as mostly a real shock, whereas the severe hit to Japanese stocks, and indeed most foreign stocks, would reflect a monetary shock, that is, the failure of central banks to keep Brexit from depressing expected NGDP growth.
Arnold Kling objects to me calling declines in expected NGDP growth "monetary shocks." Ben Bernanke and I think the term is appropriate, as they are shocks to the value of money. But even if Arnold were 100% correct, it would not have any practical implications for my policy views. If someone insists that I call then uncertainty shocks, or banana shocks, then I'll simply insist that it's the central bank's job to prevent uncertainty shocks, or banana shocks. The bottom line is that unstable NGDP causes economics dislocation, regardless of the ultimate cause of the change in NGDP. Central bank errors of omission and commission are equally bad, equally inexcusable. And central banks have near infinite ability to adjust the quantity of fiat base money, and hence have the tools required to stabilize its value, in terms of the share of NGDP that can be purchased with each dollar, or pound, or yen.

COMMENTS (13 to date)
Yaakov writes:

The news came in only three days ago. I have heard that short term spikes in the stock market are meaningless, and that you have to look at the long run, to come out with meaningful information. For example, in Israel, every time there are prospects of war, the stock market goes down, but within a week or two it is usually back up where it was before the outbreak. Do you agree, and we should read your post as meaning: "if the current level persists in the long run"? or do you disagree and believe a spike back up is very unlikely? or do you disagree and believe that even if the market goes back up next week, the current market level is the correct NGDP evaluation for now?

Thanks for the post.

There is a large real effect on the mainland through the policy channel:

The 4 large countries were Germany, the UK, France, and Italy. Now, if the UK does leave, it'll be Germany, France, Italy, and Spain.

Michael writes:

Yaakov, I don't think stocks show NGDP directly. In combination with other prices you may deduce some macfoeconomic effects, but it seems at least somewhat contentious. It's certainly not 'model-free'

As in Israel, stocks have mainly reacted to a big (but unlikely) downside risk. If it fails to materialize (as it mostly, but not always, does) they go up again. Still was right to sell while desaster loomed

Mike Sax writes:

So, Scott, with apologies, I'm just going to ask you the question which for me begs: how low does the pound go?

Mr. Econotarian writes:

I want to know what Krugman thinks about “economic policy uncertainty” of Brexit :)

This is real regime uncertainty. Will you have to send your French employee working in a brokerage in London back to Paris? Do you have to re-register your EU trademarks and patents in the UK? What tariffs, fees, and regulations will be required for goods and services trades between UK and EU? What about between England and independent Scotland (and independent Northern Ireland?) Will your Polish plumber be able to fix your toilet?

Matthew Moore writes:

Thanks a lot Scott, that's very helpful.

I think you see this as a response to a supply-side real shock, where the risk of reduced trading opportunities is identical to worsening technology. Monetary policy (as seen as a sort of reaction function) is unchanged.

Whereas previously, leaving the ERM permitted the BOE to relax the money supply restrictions that were needed to maintain the fixed exchange rate (I dimly recall the impossible trinity of free capital flows, fixed exchange rates and flexible monetary policy). Technology was unaffected. This was a move to a more favourable monetary policy. The export boom was secondary / a corollary.

PS my claim to fame is that Mark Carney and I had the same PhD supervisor, so I hope your confidence in him is justified, as I enjoy the halo effect.

Scott Sumner writes:

Yaakov, You said:

"I have heard that short term spikes in the stock market are meaningless, and that you have to look at the long run, to come out with meaningful information."

I would argue just the opposite; the short term spikes are what matters, as they show the impact of the news event. Long term moves reflect 1000s of factors, and hence it's hard to single out the impact of Brexit.

So I'd say it's the best estimate right now, although of course markets will change over time, as new information comes in.

Mike, I don't predict asset prices.

Matthew, Glad this was helpful.

Philo writes:

Scott, you wrote: "David [sc., Henderson], If there is no immediate change in the supply of money, and if expected future NGDP declined, then I'd say the demand for money probably increased, reducing velocity." But is it not true that expected future NGDP might have declined because of a REAL shock, with no change in the supply of or demand for money?

ThaomasH writes:

I agree, so why cause a lot a confusion with the nomenclature. Just recommend the policy you think is appropriate given the target you think is appropriate, everyone doing massive amounts of QE until ngdp is back on its pre-crisis trend line. The only difference is that the amounts of QE needed have probably increased as a result of Brexit.

ChrisA writes:

Of course gilt rates have dropped significantly, which also suggests deflationary rather than an inflationary aspect to this crisis. The obvious response from the BOE would be to announce a new round of QE until NGDP growth resumed it's target path. But that won't happen.

James Alexander writes:

Market-implied UK inflation didn't move much on Brexit day. Near term inflation expectations were up a bit even, longer term down a bit.
[see the last spreadsheet of the five on offer]

To be fair, market-implied inflation in the UK is not a very good guide to future inflation expectations. Amongst actuaries and financial professionals there is a well-known massive shortage of longer-dated inflation linked gilts relative to UK life and pensions' companies liability-management requirements. Those real securities are thus artificially expensive, raising the gap between nominal securities, leading to an overestimate of expected inflation.

Scott Sumner writes:

Philo, It depends how you define the demand for money. If you define the demand for money as M/P, then yes, you are correct. If you define it as M/NGDP, then no.

Mike Sax writes:

I knew you'd say that, but it was worth a shot.

Comments for this entry have been closed
Return to top