Author:Gudgin, Graham
  1. Introduction

    The result of the referendum on membership of the European Union in June 2016 generated a large shock to the UK economy. Even after triggering the formal Article 50 mechanism in March 2017 to begin the process of leaving, the final arrangements for trade and migration are not yet known. The UK government intends to achieve an exit from the EU which returns control of migration to the UK, involving leaving the single market, and removing the UK from the jurisdiction of the European Court of Justice. The aim is to secure a free trade agreement with the remainder of the EU, but if this is not feasible then the UK will leave without a formal trade agreement and rely on World Trade Organisation (WTO) rules to govern its trade with both the EU and with the rest of the world.

    The UK was already a semi-detached member of the EU, outside both the Euro single currency area and the Schengen area of passport-free movement of people, and as a result the likely impact of leaving the EU will be less of a shock than might otherwise have been the case. Even so, leaving will involve one of the largest changes in the institutional arrangements for the UK economy since joining the EU in 1973. It is not of course the only large shock over this period. The accession of the Eastern European A10 states between 2004 and 2013 represented a large shock, albeit one not immediately recognized, in setting up the large-scale immigration flows in the UK which became one of the two strongest factors behind the "leave" vote in the referendum.

    In this paper we use the Centre for Business Research (CBR) macroeconomic model of the UK economy to estimate the potential impact of what has come to be known as "Brexit." From the outset we need to say that no normal forecast is possible. The CBR model is an econometric model which uses a large set of equations to forecast future trends, each equation based on data covering the last few decades of UK economic behaviour. (1) Because this period has been almost wholly one in which the UK has been a member of the EU, the equations contain little or no direct information about how the UK would fare outside the EU. Put simply, leaving the EU is a unique event; no country has ever done this. The best we can do is to construct a series of scenarios based on assumptions about future trading arrangements, migration controls and about the short-term uncertainties which could affect business investment in the run-up to the likely leaving date of 2019.

    Our estimates of the impact of Brexit will depend partly on the nature of the CBR model and we will say a little about this. Mostly the estimates will reflect the assumptions entered into the model. Much was written and said during the referendum campaign about such assumptions, much of it highly controversial. Most detailed were the two major reports from H.M. Treasury, one on the long-term impact and the other on the more immediate consequences of a vote to leave. (2) Although the analysis in these Treasury reports was inevitably colored by the Government's stated opposition to leaving the EU, the two reports, together involving 280 pages of analysis, offered a comprehensive literature review and were based on best practice in that literature. We thus review the Treasury's methodology leading to their conclusion that a complete break with the EU Single Market would lead to a loss in GDP of 7.2% by 2030. Since the Treasury analysis strangely says little directly about the UK's trade record within the EU we also examine this in detail to see whether this supports the more indirect methods used by the Treasury in assessing the impact of EU membership on the volume of trade.

  2. The CBR Macro-Economic Model

    The main burden of this paper involves assessing what assumptions should be entered into our CBR macro-economic model and then using these assumptions to generate forecasts for two scenarios over the period 2017-25. These issues are dealt with below, but first we describe some of the relevant context of the UK economy and the way in which the CBR model approaches key issues.

    Something has gone badly wrong with economic growth in the UK where a relatively consistent growth trend of close to 2.5% per annum has comprehensively broken down (Chart 1). Similarly dramatic breaks of trend can be observed for the USA and the EU although in the latter case the slowdown began rather earlier in 2000 coinciding with the introduction of the Euro. These breaks of trend are related to the so-called "productivity puzzle" for which economists have no agreed explanation. The puzzle is to explain why the growth of output per hour worked has slowed since 2000, and especially since the financial crisis of 2008. Alongside the failure of existing forecasting models to predict the 2008 economic crisis this break of trend provides another reason for developing a new model which can predict and help to account for these bewildering trends. Our general view is that the slowdown in growth is due to credit conditions in a post-crisis world with a badly impaired banking system. Perverse government austerity programs in major economies have exacerbated the situation but the main cause is financial.

    2.1. Consumption, borrowing and credit super-cycles

    One key feature of the model is the important role of credit in generating business cycles. The consumption function shown in Table 1 has conventional features in that consumption depends on disposable income and wealth. Importantly, these loans are taken out to purchase houses (excluding remortgaging) but around 75% of the loans are for the purchase of existing rather than new dwellings and these are thus loans which end up largely as bank deposits of those selling houses (often inherited property). The evidence of the equation above is that a proportion of these deposits are used to finance consumption.

    This in turn is important because of the volatility of mortgage credit. The number of housing loans has fluctuated in large 20-year cycles, termed financial cycles by Claudio Borio of the Bank for International Settlements, to distinguish this long-period financial cycle from the usual business cycle of 3-5 years. We will use the term super-cycles to emphasize the long-duration of this phenomenon. The extended period with a very low volume of loans since 2008 is unprecedented in the post-war economic history of the UK and is largely responsible for the sluggish growth of GDP over this period. This is the way in which a badly impaired banking system prevents a normal recovery from a deep recession. Our estimate is that the potential demand for loans is currently at historically high levels due to very low mortgage interest rates, but the number of loans is low due to banks' restrictions on the supply of loans including requirements for sizable deposits. With house prices remaining very high in the UK, the requirement for substantial deposits places a large barrier in the way of new buyers.

    The importance for this in assessing the impact of Brexit lies in the context it sets for economic growth. Credit is currently on the upswing of the latest super-cycle leading to reasonably rapid rates of household spending. This upswing, helped by government schemes to stimulate house purchase for first-time buyers, allowed the previous Chancellor of the Exchequer, George Osborne, to pursue a policy of mild public sector austerity without doing much harm to the growth of aggregate GDP. A continuing upswing for the next five years would provide a favorable context for the disruptive process of leaving the EU. Beyond the middle years of the next decade we had expected before the referendum that the credit cycle would turn down, as demand for loans became the main constraint on loan volumes with demand depressed by high debt levels and falling real wages. Chart 2 shows that the cycle is now expected to continue its sluggish recovery, towards fully meeting demand for housing loans which is potentially large when interest rates are as low as they have been in recent years.

  3. Assumptions on Brexit

    The difficulty in generating any forecast for the future of the UK economy is in knowing what to assume about both future trade arrangements and the short-term impact of uncertainty about these arrangements. As we have stated, the best that is possible is to generate scenarios based on assumptions about these things. This is not to say that there is little on which to base assumptions. A plethora of reports were produced during the referendum campaign to assess what the impact might be of a vote to leave the EU and, several months on from the referendum, some consequences have also begun to emerge.

    3.1. Short-term impact of Brexit

    These reports published during the referendum campaign generally produced separate estimates for both the short-term impact of uncertainty and the long-term impact of changed trading arrangements. A summary of short-term impacts from non-government sources is shown in Table 2. The government's own estimates are shown in Table 3. The estimates vary depending on what is assumed about the nature of the likely eventual relationship sought with the EU. In general the largest estimates of losses of GDP stem from an expectation that the UK will leave the single market and customs union and will fall back on WTO rules. Something of a consensus emerges from these studies with an expectation that uncertainty will reduce GDP (relative to a pre-referendum baseline) by around 1% after one year, 2-4% after 2 years, 3-4% after three years and 4-6% after 5 years. The Treasury's estimates are at the high end of this spectrum of views with a view that GDP would be reduced by between 3.5% and 6%.

    The Treasury summarized its own view in the following words, "The analysis shows that the economy would fall into recession with four quarters of negative growth. After two years, GDP would be around 3.6% lower.... the all in the value of the pound...

To continue reading