Did Brexit Really Shrink the UK Economy? - Anonymous Banker

Brexit Myths Debunked (Volume II)

I have a friend whose economic and financial analysis is always interesting because it is apolitical. He is a former City economist with decades of experience. Unfortunately, like many people who have something genuinely valuable to add to the conversation, he prefers to protect his privacy in our turbulent political and cultural landscape. But I’ve always found his explanations and insights useful and I think you will as well. That’s why he’ll be contributing to my Substack from time to time as the Anonymous Banker.

This piece is part of a series of articles which attempt to provide an objective, data-based look at the economic impact of Brexit on the UK and debunk the myths peddled by both sides of the debate.

Read Volume I here.

---

Former Bank of England Governor Mark Carney recently made the following comment to the Financial Times that was then gleefully picked up by Remain supporters:

“Put it this way, in 2016 the British economy was 90 per cent the size of Germany’s. Now it is less than 70 per cent.”[1]

Whichever way you voted, there is no pretending this isn’t a heavyweight intervention in the Brexit debate. Mark Carney has impeccable economic and financial credentials – ex-Goldman Sachs, BA in Economics from Harvard, MPhil and PhD from Oxford before joining an elite cadre of central bankers having presided over the Bank of England and Bank of Canada. His comments therefore are very important.

Which is why it’s such a shame he is talking nonsense.

A useful rule of thumb for the major European economies over my time in finance has been that France and the UK are roughly the same size at any given time, and that the UK is roughly 2/3 the size of the German economy.

Let’s look at the numbers and see for ourselves. Carney was intellectually honest enough to present the numbers to support his assertion via Twitter:

Mark Carney @MarkJCarney

Sure. Jan 1 2016 (pre referendum call) 🇬🇧 £2.10 trn or 94% of 🇩🇪 £2.2 trn. Now 🇬🇧 £2.17 trn or 70% of 🇩🇪 £3.10 trn. @TimHarford @FT

David Anderson @bricksilk

@CJCHowarth @MarkJCarney @FT Hmm. Can you show us the basis for your very striking statement to @FT, @MarkJCarney? Worth a look on More or Less @TimHarford?

Looking at the figures he provided, everything seems to check out. But look again.

For starters, if the size of the German economy had grown from £2.2Tr to £3.1Tr during this time period that would be a cumulative increase of 41% in Sterling terms! This would be equivalent to annual growth in real GDP of 5.8%, which in a major advanced economy is unheard of, with circa 1.5-2% being the norm.

Carney’s analysis is incorrect not only mathematically, but also logically.

Let’s use a neutral and respected source: the IMF World Economic Outlook Database. The IMF shows UK GDP of £2.09Tr (i.e. £2,089.3bn) in 2015 and £2.27Tr in 2022, which is essentially in line with his figures. We also agree with Carney’s figure for German GDP in Sterling terms in 2015 of £2.2Tr. Alles gut so far….However, his figure for German GDP in Sterling terms in 2022 looks like an error, and quite a serious one. I calculate the correct figure as being £2.77Tr, not £3.1Tr as shown in the table below.

Source: IMF World Economic Outlook Database, Bloomberg.

Ah! But Anonymous Banker, you said “Sterling terms” above – wouldn’t the decline of Sterling explain this? Average Sterling-Euro FX rate in 2016 was 0.8193 and 0.8526 in 2022. That’s a depreciation (ie. It costs you 85p to buy a Euro now, compared to 82p in 2016; Depreciation = costs you more in £) of around 4%, a long way from the 41% his figures imply.

The IMF states that Germany’s GDP in 2021 was EUR3.203Tr and is estimated as EUR3.253Tr in 2022. Using the average Sterling-Euro FX rates for 2021 and 2022 gives us a German GDP in Sterling of £2.754Tr in 2021 and £2.773Tr in 2022. Which is very different to his £3.1Tr.

It would appear that to get to his £3.1Tr figure, Carney has taken German GDP at current prices (i.e. unadjusted for inflation) and then converted it to Sterling at current FX rates. German GDP in 2021 at current prices per the IMF was EUR3.601Tr. Multiply this by the average Sterling-Euro rate for 2021 of 0.8597 and you obtain £3.095Tr, or circa £3.1Tr. QED.

However, Carney’s other figures are all in real terms (i.e. adjusted for inflation), as they should be for this sort of analysis. He really is comparing apples with oranges, which renders his analysis flawed.

It is also odd that he chose to use current exchange rates to convert German GDP into Sterling for comparison. Exchange rates are notoriously volatile. Basing your analysis on one year’s data, using the relevant exchange rate at the time is likely to lead to very different results depending on the time period you pick. Let’s illustrate this, and Carney’s error, by looking at the real data.

Source: GDP Data from IMF World Economic Outlook Database, FX Rates from Bloomberg.

It’s clearly logical to take GDP at end 2015 (i.e., pre-Brexit) as your base year and compare it to today. But you also need to be sure you’re comparing like with like (which he wasn’t as shown above), and you also need to be careful you haven’t picked an outlier as a year to make your comparison from[2].

As the table shows, if we look at 2015, Carney’s point of the UK economy being 90% of that of Germany’s is borne out. However, looking at the full time series of data one can see that 2015 is very much an outlier compared to other years. What drove this figure to 95% in 2015? Ironically, the value of Sterling relative to the Euro was extremely strong compared to history in that specific year – 1 Euro cost you 72.6p in 2015, compared to an average of 84p between 2010 and 2022.

Therefore, if we were to follow the Carney logic, but correct for these errors, it would be fair to say that in 2015 the UK economy was 95% of the German economy, compared to 82% now (versus his claim of 94% to 70%). The data above shows how representative this figure of 95% is, and the reader can judge for themselves whether it is really reflective of the two economies. Excluding the 95% figure, the average size of the UK economy to Germany has been 79.8% - which is where we stood in 2021, and compares to 82% in 2022. Hardly the relative decline Carney has argued, is it? Even using his flawed methodology.

Carney’s other error in using exchange rates to make GDP comparisons is that currency markets can see large swings in very short periods. These moves are often for reasons wholly unrelated to economics, and more to do with speculative forces. For example, the Sterling-Euro rate varied by 9% from high to low in 2022. This means that your choice of conversion rate could lead to one of your countries’ GDP being nearly 10% bigger or smaller relative to the other. This is a huge margin of error given these economies typically grow 1.5-2% annually.

So, this is clearly not an exact science. In order to make more meaningful comparisons, economists tend to look at the trend over time (to avoid cherry picking a base year that suits your argument) and in the particular context we are discussing, they use what are called Purchasing Power Parity (or PPP) exchange rates.

PPP attempts to capture the fundamental value of one currency in terms of another. The best way to understand this is to use the wonderful creation of The Economist magazine – The Big Mac Index. Simply put, PPP asserts that the cost of an item in London should be the same as that in New York when adjusted for currency. Therefore, if we take the cost of an item in London and compare it to that in New York, the ratio ought to give us the relative value of one currency in terms of another. The Economist chose to use the McDonald’s Big Mac given its global availability and broad basket of input ingredients (beef, bread, salad, packaging etc). Other examples might be a can of Coca Cola or furniture units in the Ikea catalogue.

Why should this relationship hold? If a tradeable good is overvalued in one country and relatively undervalued in another, commercial actors will look to source the items in the cheaper country and sell them where they are more expensive. This increases supply in the expensive country reducing prices, whilst the exporting of the good leads to less supply in the cheaper country, raising its relative price.

Source: https://www.economist.com/big-mac-index.

Please note the 0.81 quoted above means it costs 81p to buy a dollar. The cost of a pound in dollars is $1.23 as mentioned above, and the way you will see it quoted in the media. The two numbers are the inverse of each other – 1/0.81 = 1.23.

Economists prefer to use PPP for this sort of comparison because it is a much more stable measure than exchange rates and has a much more theoretical and logical anchor. Exchange rates will deviate wildly from those implied by PPP in the short term but in the long term, PPP is a good indication of the relative value of two currencies.

So, what does PPP have to say about the point Carney made?

Thankfully, the IMF database provides us with a measure of the UK and German economies on a PPP basis. The table below shows this data and the size of the UK economy relative to Germany’s over time:

Source: IMF World Economic Outlook Database

On a PPP basis the UK economy in 2022 was 71% the size of Germany’s. This compares to 71.9% in 2015 and 70.3% in 2016, and its average over 2010-22 of 70.1%. Does that strike you as a huge relative decline?

Other economists and commentators have called Carney out on his claim and he sought to justify his comments on BBC Radio 4 in the days following his interview. He makes some important points which I believe are borne out by the data, and others that are not.

With regard to his central claim regarding the impact of Brexit on the size of the UK economy relative to that of Germany, Carney is quoted as saying:

“The question is the purchasing ability, the international weight of the economy, which was [pre-Brexit] at a different level relative to the German economy, relative to the Canadian economy, relative to other economies than it is today.”

This assertion can again be verified by the data. The IMF provides a data series which shows the size of an economy as a percentage of the total world economy GDP, on a PPP basis. This is an ideal dataset to examine his thesis.

In 2015 (ie pre Brexit), the UK represented 2.499% of the world economy. By 2022, this proportion had shrunk to 2.333%. Case proven for Carney?

Let’s take a look at the same statistic for the US and other major European economies:

Source: IMF World Economic Database. Series: Gross domestic product based on purchasing-power-parity (PPP) share of world total

The IMF data shows that all major European economies now represent a smaller share of global GDP than they did in 2015. The UK’s loss of share is marginally larger than that of France, and, ironically, smaller than that of Germany. Brexit also appears to be Kryptonite for the US, which lost 0.8% share of the world economy over this period.

Carney made some further claims which are accurate and an unpleasant truth Brexiteers have to reckon with:

“Sterling moved against all major currencies from the point at which the referendum was called, and then it moved more sharply after the referendum result. It hasn’t recovered. It’s fluctuated around but it has not recovered.

This is what we said was going to happen, which is that the exchange rate would go down, it would stay down, that would add to inflationary pressure, the economy’s capacity would go down for a period of time because of Brexit, that would add to inflationary pressure, and we would have a situation – which is the situation we have today – where the Bank of England has to raise interest rates despite the fact that the economy is going into recession.”

The broad thrust of his argument here is correct: the true measure of the world’s opinion of the UK currency is captured by its exchange rate to the US dollar. The pound has fallen from $1.4877 prior to the Brexit vote to around $1.20 currently, a 19% decline. Make no mistake about what this means – our collective wealth in the UK buys less assets in US dollar terms than it did prior to Brexit. These moves are not unprecedented, however: Sterling fell by 31% in 2008/09 and by 15% in 2014.

It is important to point out that currency valuations are two sided – another key driver in the exchange rate is dollar strength, as well as the relative unattractiveness of the pound. Investment professionals gauge the strength of the dollar using a so-called Dollar Index, which captures the value of the dollar relative to a basket of the world’s major currencies. This measure has risen some 8% since Brexit. Contrary to what you may have read in the UK media about our battered currency, it may surprise some to learn that the pound actually recovered to pre-Brexit levels in 2018 and 2021 at $1.42. It has also reached a Truss-induced low of $1.035. For additional context, it’s also worth noting that the euro went through parity (i.e. it cost less than $1 to buy €1) against the US dollar in Q4 of 2022.

Carney is correct to suggest that the weakness of sterling will have fed into UK inflation. Most input commodities (eg. metals and oil) are priced in US dollars. He’s also quite correct to say that the Bank of England has been forced to raise rates despite the weakness of the economy. However, this is a strangely UK centric view – the inflationary problem we face is present across the G7 and has resulted in the Federal Reserve and European Central Bank having to raise rates aggressively. So, again, many of the issues our media commentators bemoan are actually also true of our peer G7 countries.

Sadly, given his profile, Carney’s erroneous comment has been widely reported and accepted as gospel. Truth is the first casualty of war, and the battle over Brexit is no exception.

Brexit Myths Debunked (Volume III) will be published in the next couple of days.

---

[1] Mark Carney, Lunch with the FT, Financial Times, October 14^th 2022. https://www.ft.com/content/f1f0a66a-fa2c-4d70-9874-8003bdb3fb53

[2] This is like Priti Patel claiming a reduction in theft in 2020 versus 2019, whilst ignoring there might be a very good reason why there were fewer people shoplifting in 2020…. https://www.indy100.com/news/priti-patel-coronavirus-briefing-shoplifting-crime-rates-9484351