David R. Henderson  

Is Brexit a Monetary Shock?

One Cost of the Drug War... Weekend Grab Bag...

Both co-blogger Scott Sumner and blogger David Beckworth argue that the Brexit vote is a large monetary shock. To be a monetary shock, it has to be either (1) a shift in the supply of money or (2) a shift in the demand for money.

Scott argues that it's not a shift in the supply of money. That then leaves only one possibility: a shift in the demand for money. But neither Scott nor David Beckworth makes the actual case that would make that clear.

How do we know there was a shift in the demand for money? And which money? The euro? The pound? Both?

They may well be right but, after having read their posts 3 times, I don't see it.

Comments and Sharing

CATEGORIES: Monetary Policy , Money

COMMENTS (25 to date)
Roger McKinney writes:

Seems that he means that the drop in stock markets indicates a rush to cash. Could be, but I'm betting the drop is temporary. No one should take a single day move in the stock market to mean anything at all.

Arnold Kling writes:

What's a matter, son? Can't you see the emperor's clothes?

Seriously, asking what sort of money is being demanded, in which countries, is a good question. If you look at currency markets, an appreciation of the dollar suggests a stronger demand for dollars. If you look at which stock markets took the biggest hit, then the biggest increase in money demand was in southern European countries. Seems hard to fit into an old-fashioned monetarist perspective.

Actually, I think I could pass a Turing test as a market monetarist with the following 4 statements:

1. An increase in money demand is equivalent to a decrease in velocity.

2. The stock markets are saying that future NGDP will be lower than otherwise because of Brexit.

3. Sticking to the same path of money growth, this means that velocity will be lower than otherwise.

4. Ergo, this is a shift in money demand.

To me, (1) - (4) constitute an intellectual swindle. It is a trick used to label any and every shock to NGDP as a monetary shock.

Market Fiscalist writes:

The most obvious economic effect of Brexit would be increased barriers to trade. This seems like it would be a demand shock - suppliers face less demand at any given price because of increased transaction costs.

Khodge writes:

As best as I have been able to determine, nothing of substance, financially, has really happened. They've curtailed, apparently, the EU technocrats which, as a rule, can only mean good things. Since Britain is not on the Euro, no change in money supply happened.

If the technocrats choose to punish Britain by raising tariffs, that really becomes a different question. In some sense, this whole thing sounds like the sequestration which, despite the technocrats' protest, did wonders for the US economy.

Craig Pirrong writes:

I'm glad it wasn't just me. The logic was inductive (from the observed market price movements) not deductive (and then tested on market price movements).

It seems to me that this is a shock to uncertainty/risk. That has implications for asset prices, options prices/implied volatility, money demand, investment, etc.

Justin Rietz writes:

I'd argue that money is the transmission mechanism, but not the source, of the shock. Given that we don't live in a barter economy, this is unavoidable, and I believe important to recognize particularly when we consider monetary policy. Therefore, I don't think it's accurate to call it a "trick".

Steven Horwitz writes:

"Does "never reason from a price change" apply to exchange rates too?" - Jeremy Horpedahl on Facebook today.

Bob Murphy writes:

Good catch, David. Naturally, I'm not the most neutral arbiter, but I think this is another example where Scott's analysis *could* be right, but he assumed his conclusion without actually offering evidence for it. (Analogous to what happened when he was talking about Russia and you pointed out the potential problem.)

So, quoting from Scott's post, I think he's thinking like this:

1. Plunging stock prices
2. Plunging bond yields
3. Plunging commodity prices

This is how global markets react when NGDP expectations fall. In contrast, negative real shocks are inflationary.
2. A monetary shock only depresses RGDP because of a decline in NGDP (and/or inflation) expectations.

So, I think Scott is saying that he is quite sure from the various price movements that there was a fall in expectations of NGDP growth (even though we can't directly observe this, since we lack a liquid market in NGDP futures), and then--by definition--this translates into monetary tightening, in Scott's framework. We can likewise expect real growth to be lower than it otherwise would have been, due to the passive tightening.

Jim Rose writes:

Real business cycle theory explains it all.

Stephen writes:

No economist I, and I don't know about monetary shocks, so allow me to take a different tack.

Although it is unprecedented, perhaps Europeans can manage Brexit by studying the history of nonviolent decolonization. When nations became independent, they had to create new institutions (or build up existing ones), laws, and systems, often hastily with little training. The transition to nationhood was rocky and usually needed injections of outside capital.

The British are light-years ahead of former colonies who are just spreading their wings. Great Britain can draw upon centuries of institutional knowledge and a highly educated civil service as it re-acquires full sovereignty. Besides, the descendants of the greatest Empire in history should know how to do independence, and independence (India, Eastern Europe) is regarded as a good thing for the world, right?

Toby writes:

@Bob: aren't 1-3 evidence of a monetary shock? Sure these could also indicate a real shock, but real shocks would be negatively associated with inflation expectations, whereas monetary shocks would be positively correlated with inflation expectations. If it is both it depends on the magnitude of each what the net effect is I suppose. Here it seems that the monetary shock dominates.

I think you can, therefore, distinguish a monetary from a real shock here is by asking what else would be consistent with the same goods being chased by less money and is this different from fewer goods being chased by the same money. Looking at inflatiom expectatioms should provide a test to distinguish one from the other. Right?

ThaomasH writes:

I do not think Scott is wrong within his framework to call Brexit a "monetary shock" but the needless dispute over terminology is not helpful unless it is linked to policy. The Fed and even more the ECB should be buying stuff like mad until at least the price level (if not NGDP) is back on its pre-crisis track.

Bill Woolsey writes:

Sumner's argument is that a negative real shock shifts the aggregate supply curve to the left and so raises inflation expectations. This should cause bond prices to fall and commodity prices and while stock prices could go either way in theory, they frequently rise. That didn't happen.

A negative demand shock causes the aggregate demand curve to shift to the left. This should reduce inflation expectations. This should cause bond prices to rise and commodity prices and stock prices to fall. That did happen.

Negative demand shocks occur when either the quantity of money falls or else the demand to hold money rises. Both are monetary.

Beckworth explicitly claims that there has been an increase in the demand for dollars--as evidenced by the increase in the dollar exchange rate. This increase in the value of the dollar also raises the value of currencies tied to the dollar.

Jody writes:

Why can't it just be a risk shock which influences all kinds of other behaviors? In this case, take actions to reduce risk which can reduce demand AND supply while also monkeying around with various monies.

Bob Murphy writes:


I think what David is getting at, though, is that the pound fell sharply against other major currencies. So it would be weird to say that the fall in British asset prices is due to an increased demand for pounds.

The pound also fell pretty sharply against the euro. So is it an increased demand to hold euros? If so, then is that equivalent to a monetary loosening on the British side? If so, then why did British asset prices fall?

I'm not saying the story is impossible, but it's a lot more complicated (I think) than Scott's post suggested. I think he was starting from the perspective of a single currency, and asking what would make all those prices fall. Ah, an increased demand to hold money (coupled with sticky prices/wages)!

But when you're trying to explain why British, German, and US stock prices fell, while the British pound fell sharply against the other two currencies, it gets a little trickier to explain that as "increased demand to hold money."

Scott Sumner writes:

David, What Arnold calls a "swindle", I call a useful definition. I define the stance of monetary policy in terms of expected NGDP growth.

Bernanke once offered a very similar definition.


Perhaps Bernanke is also a swindler, but I don't think so.

Of course something can be both a monetary and real shock, and that's why I often suggest the counterfactual of what would happen of the Fed pegged the price if NGDP futures contracts.

Scott Sumner writes:

Steven, Yes it does apply to exchange rates. Before drawing any implications from a change in the exchange rate, you need to have some idea as to why it changed. That's why event studies are so useful.

Thus yesterday, it's not plausible that the yen soared in value because Japanese productivity soared (good news) but it is plausible that the yen soared because of more demand for the yen as a safe haven currency (bad news) The concurrent steep plunge in Japanese stock prices confirms that view.

Scott Sumner writes:

Bob, My comments applied to the global economy. The steep fall in the British pound is certainly consistent with a negative productivity shock hitting the UK.

But at the global level, I see clear evidence of a fall in NGDP expectations.

Jose Romeu Robazzi writes:

This a very rare event in the world economy, and it is very strong evidence supporting Prof. Sumner views. We have an event, markets reacted strongly, although nothing really happened yet! There is no best example of long and variable LEADS!

So it MUST be something about expectations that end up reflecting higher demand for money, anything resembling a safe haven, at least monetarily. On average, all cryptocurrencies went up sharply, gold went up sharply, yen, treasuries.

I think it is much more difficult when there is no event. But even now, what the hell have japanese equities to do whith UK leaving the EU ??? If it wasn't for the Brexit event, and stocks fell around the globe (as they have without apparent reasons many times beofore), economists would be um much more dire straits, and fighting over nothing. Thanks god for the Brexit event, so we can reasonably argue for a monetary shock without receiving blank faces in return ...

BC writes:

@Scott, to determine whether the shock in UK was primarily monetary or real, would we look at British yields and perhaps commodity prices in GBP? If British yields and commodity GBP-prices were rising, indicating rising inflation expectations, then the shock would be mostly real. If yields and prices were falling, then the shock would be mostly monetary?

David R. Henderson writes:

@Scott Sumner,
David, What Arnold calls a "swindle", I call a useful definition. I define the stance of monetary policy in terms of expected NGDP growth.
But I wasn’t asking about the stance of monetary policy.
Here’s the question I asked:
How do we know there was a shift in the demand for money? And which money? The euro? The pound? Both?

ThaomasH writes:

And I still want to know what are the policy implications of knowing that Brexit is a "monetary" shock (as opposed to thinking that it is a belief that demand for C, I and maybe even G will fall as a result of the uncertainty) And if you say that it is the same thing, don't b;me people for thinking you have a queer view of "monetary."

What should the Fed do? Buy assets or sell them? and which ones and how much?

Scott Sumner writes:

BC, I suppose you could look at TIPS spreads, as well as various proxies for RGDP growth. I wish we had a NGDP futures market, Brexit would have provided a wonderful natural experiment.

David, 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.

Thaomas, The Fed should do enough QE (or reduce IOR by enough), to keep NGDP expectations from falling.

Thomas Strenge writes:

I'm with David, I don't see what the problem is. The politicians tried to scare people into believing that the world would end with Brexit in order to coerce them into voting EU. Now that Brexit has happened, plenty of fools must act accordingly. There is a danger to free trade, but it is unlikely. The British people support free trade and Europe would have more to lose from a trade war than the UK. Indeed, the Germans are already proposing associated EU membership which would continue all free trade agreements. I am sure something will be worked out. The biggest losers appear to be British banks who will no longer benefit from easy ECB money. At least some of the central bankers at the Bank of England have read Bagehot and are unlikely to be as profligate as their colleagues in Frankfurt.

Brian Goff writes:

Has there been an increase in the demand for the money (the dollar; equivalently a decrease in velocity)? Not sure how this is even at issue. An increase in FX value of the dollar would seem to indicate an increase in demand for $ relative to other currencies. A decrease in Treasury rates suggests an increase in demand for highly liquid, dollar denominated assets (money, at least for institutions) relative to other kinds of assets. This requires thinking of "money" outside of a 1970s/1980s M1 or M2 box.

Why the shift? This is where Craig Pirrong's comments seem on the mark. Is it that NGDP expectations (along with earnings ...) have fallen or that discount rates have risen due to increased uncertainty. The indirect evidence (all the "stochastic discount rate" stuff) would say that such large short term moves are due to discount rate moves much more than expected NGDP (earnings, ...). David Beckworth uses the "rush to safe assets" terminology, but it's the same thing, I think.

Comments for this entry have been closed
Return to top