Winner of the New Statesman SPERI Prize in Political Economy 2016

Tuesday 7 February 2017

Does the Bank’s latest forecast mean Brexit has had no effect?

The focus of many (not all) journalists on GDP growth was in evidence again in the reporting of the Bank of England’s latest UK forecast.

Bank of England Inflation Report February 2017 and (in italics) my estimates
Growth rates of
Ave 98-07
Household Income
Savings ratio [1]
GDP per head
2.25 [2]
Household Income p.h.
[1] Level [2] Average 1955-2007

The headline news was Brexit didn’t seem to be having much effect on GDP growth, despite earlier pessimism from the Bank. Leavers have never forgiven the Bank for giving their pessimistic views on the immediate impact of Brexit during the campaign. There is also an attempt to suggest that because many macro forecasters have been surprised by the resilience of the economy since the vote that must mean that the near universal view of economists that Brexit will be bad in the medium term is also now very suspect. Anyone who knows about these things knows that an unconditional macro forecast is very different from a conditional forecast based largely on international trade evidence, but as most people do not know these things (including most political journalists) it is an effective bit of propaganda.

A major reason for the more optimistic forecast now is that consumers so far have decided to reduce their saving, which the Bank had not expected. One possible reason for this is that a lot of consumers have decided to undertake major purchases like buying a car to beat the coming price rises expected as a result of the depreciation. That alone would imply that the decline in the ratio is temporary, but as we shall see, the Bank is now expecting it to continue.

It is hard to forget a remark made to a fellow economist during the referendum campaign by the member of the audience in a public meeting in the North West. After this economist had talked about the beneficial effects of joining the Euro on GDP growth, they said something like ‘it may have helped your GDP but it hasn’t helped mine’. In that spirit I want to make two points that were generally ignored by the media in their reporting of this forecast.

First (as regular readers will know), GDP is the output of the country, not the output of an average member of that country. Although the ONS now releases estimates of GDP per head (or per capita as it is often known) with its GDP estimates, most journalists seem to have not noticed. One reason for the focus on aggregate GDP is that forecasters like the Bank continue to publish only aggregate figures.

The table above is an attempt to adjust the Bank of England’s forecast for expected growth in the population. I’ve basically just taken the average population growth rate for the last few years and projected it forward. That could be on the high side if immigration from the EU falls off substantially over the next few years, but this would probably only increase the numbers by another 0.25%. Growth of 1% in GDP per head does not sound so good, particularly when you note it is less than half the historic average.

Second, over the following few years even GDP per head is likely to not feel like ‘my GDP’. We can see this from the Bank’s forecast for real household income. These fail to get above 1% growth. The reason is something Leavers do not like to talk about, and which therefore many journalists ignore: the impact of the Brexit depreciation in sterling. As this depreciation gradually leads to inflation not matched by higher nominal wages, real income growth will suffer.

Why is forecast GDP growth so much higher than income growth? The Bank now expects consumers to reduce their savings to unprecedentedly low levels. Why would they do that? The Bank operates a model where consumers base their current consumption on anticipated future income, and where expectations are rational. If, as economists universally expect, Brexit leads to slower income growth in the future, consumers should have reacted to that by cutting current consumption because their future income will grow more slowly relative to pre-Brexit vote expectations. This they clearly have not done, in part because many of them do not believe future income growth will be reduced by Brexit. That is at the heart of the recent forecast revisions. But this leaves the Bank without any guide to how the savings ratio will evolve. If consumers continue to believe everything is OK, despite the short term fall in their income growth, then further falls in the savings ratio are possible.

Of course even these numbers for household income are also inflated by likely household growth. (They measure all income going to households, not the income of an average household.) The final row adjusts for the expected growth in households. The average household size has remained constant over the last decade, and I have assumed that continues. As you can see, the income of the average household is at best going to be flat, and may fall slightly. So to say, as some Leavers have, that this forecast suggests Brexit will have no effect before we leave is completely wrong.

So how is the score in the match going between Leavers and economists, where goals are actual events rather than clever soundbites. The last time we looked the Leavers had let in two goals: the depreciation in sterling immediately after the vote, and then the Bank having to bring rates down to their lower bound again and start another round of QE. Nothing since then suggests those goals were invalid. If I’m feeling generous I’d say having to revise up a forecast could count as a shot on goal, but as it reflects a mixture of policy and consumers saving less I think it is also a miss.

But there has been a new goal scored by the Leavers, but unfortunately in their own goal. It is now clear to those not dependent on Brexit propaganda that as a result of Brexit we are going to be a supplicant to probably the most dangerous and right wing US president ever. Truly awful. Should never have been president. Got 5 million less votes, and that’s not counting those stopped from voting. Complete fluke. Only got the votes he did because of a biased media and fake news. FBI is an absolute disgrace. Everyone agrees. He’s got to go. Some people are saying he is unstable. Others that he is a stooge of the far right. Impeach the guy. Truly awful. (Sorry, couldn’t resist)

Which all means, the score is now
Economists 3 Leavers 0

and we haven’t even reached half-time yet. But there may be a lot more goals to come when the negotiations conclude. If the economists keep scoring, we can avoid extra time, which is desirable because that only ends in 2030!


  1. I'd be interested in hearing more about the "international trade evidence" used in Brexit forecasts.

    1. Then please read the many estimates of the long term impact of Brexit. You could start with

  2. A great read as ever Simon. I stumbled across this in the Guardian this morning and wondered if this can be squared with the arguments you make here and elsewhere, or whether it is a bit of a 'feel good' report designed to keep the spirits up in the face of some troubling times ahead (to put it mildly):

  3. I agree with your analysis here; the aggregate figures do not give a true picture of the situation.

    I believe that the position is likely to be worse than this because there is no recession built into the forecast and how likely is this in the next two years? We are, arguably overdue one now.

    And, although the savings ratio is shown as declining, does this mean that debt will not increase and that a decline in the SR and an increase in debt are mutually exclusive? I say this because I wonder what the effect of the debt burden will be in holding back consumption and whether this is reflected in the forecasts; putting it another way will the debt burden reduce growth even more?

    As far as DT is concerned you may be right but, to me, he has one big advantage. I believe that the current economic system we have is unreformable and vested interests, especially in the US, will never do this. DT is far more likely to blow the whole thing up and this is the only way it will be reformed. His suspension of Dodd- Frank is one indication that he's prepared to unleash the bankers yet again and they in their greed and venality will crash the system - again - and that, in my view, is a necessary condition for true reform.

    1. I have never understood that by taking many steps backwards, we will subsequently be able to make even more steps forward.

    2. Indeed but I was suggesting that these might be unintended collateral effects of Trump's policies, not deliberate actions on his part to that end.

  4. As this depreciation gradually leads to inflation not matched by higher nominal wages, real income growth will suffer.

    This is where Simon's anlysis breaks down or some might say falls apart.

    Simon seems to think that U.K. importers can keep raising their prices as our wages fall even though we all only have a set amount of income every month to spend on imports.

    Simon thinks we are all going to rush out and borrow money from the bank so we can all buy these imports. What he fails to see is that we won't.

    We will buy cheaper alternatives instead which only highlights the growth in cheaper supermarkets over the past few years.

    Then these importers will be struggling to keep their market share that they fought so hard for over the last 50 years. So they will have to reduce their prices and the floating rate will adjust to its true level.

    As any central bank report shows quite clearly. Pass through inflation from a drop in a floating exchange rate is minimal. Infact, If Simon could point me to any economic paper that shows a fall in a floating exchange rate leads to pass through inflation then I wish he would. He could post it on here.

    Unfortunately, it is a theory all based on using fixed exchange rates.

  5. It would be fascinating to see your analysis split by income deciles.

  6. Do you mean nearly 3 million votes Trump lost by, not 5 million?

    popular vote (CNN politics)

    trump 46.1% 62,979,879

    clinton 48.2% 65,844,954

    1. As I thought was clear, I was at that point writing in the manner of Trump, who you may have noticed has a slight tendency to exaggerate when it suits him.

  7. "consumers should have reacted to that by cutting current consumption"
    Really? Doesn't your model suggest that monetary easing leads to higher spending? Carney eased monetary policy, AD rose.

    1. True, but in most models that would be dominated by a fall in expected income. But you are right - its complicated.

  8. Good to see you admitting that 'expectations' is BS in a roundabout way.

    "Got 5 million less votes"

    Nobody knows who would have won had the system not been based on the EC. The dynamics were different. If there is a constituency system - as there is in a presidential poll then the 'popular vote' is meaningless. If you're a trump voter in a democrat state then it's like being a Tory in Barnsley. You're wasting your time.

  9. I take it the Trump par is intended in part as a humorous representation of your exasperation? Got 5 million less votes? I thought it was 3 million. Biased media? I wasn't in US at the time (or any other time for that), but got the impression that the mainstream media was overwhelmingly in agreement that Trump was a dangerous, narcissistic, unhinged clown.

    1. I was taking Trumps tendency to exaggerate e.g. biggest ever EC win. Media includes Fox News etc.

  10. The BoE having to revise its forecasts because misinformation and poor reporting mean consumers don't know that the rational response to the coming storm is to increase their savings ratio, is rather a pyrrhic victory for Leavers. (Yes, I know you didn't count it as a goal, but they did.) Funny the media didn't report it that way, either...


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.