The Impact on Brexit on UK Firms Nick Bloom (Stanford), Phil Bunn (Bank of England), Scarlet Chen (Stanford) Paul Mizen (Nottingham), Pawel Smietanka (Bank of England), Greg Thwaites (Bank of England) Preliminary – Columbia, March 5 th 2019 Disclaimer: Any opinions and conclusions expressed herein are those of the author and do not necessarily represent the views of the Bank of England.
Brexit key dates 1. June 2016 referendum, 52% voted leave. Cameron resigned as Prime Minister, succeeded by Theresa May. 2. Sept 2018 the “ Chequers plan” is rejected by EU leaders in Salzburg, raising likelihood Britain leaves with no deal 3. March 2019 Britain (currently) needs to leave (Article 50)
Brexit vote (demographics similar to Trump) Supported by older, less educated, poorer, rural voters Source: The Guardian https://www.theguardian.com/news/datablog/2016/jun/24/the-areas-and-demographics-where-the-brexit-vote-was-won
Following Brexit vote a Bank-Nottingham- Stanford team rapidly started a new firm survey • Piloted survey and questions July and August 2016 • Sept 2016 rolled out - randomly sampled population of all 45K UK firms with 10+ employees inviting them to join the monthly Decision Maker Panel • So far 8K firms in the DMP
Key findings Opinions: A) Brexit seen by firms as an massive second moment (uncertainty) shock with added first moment impact B) Firms with more prior EU exposure - exports, imports, workers and regulations – hit with greater uncertainty Regression results: A) So far…Brexit reduced employment growth by about 0.5% and investment growth by about 5%. B) Reduced productivity by about 1% (so far) from greater misallocation (likely negative within firm effects too)
Estimating the impact of Brexit is not easy – growth picked up in 2016 – hence use micro data Business Investment in G7 Economies (In percent, Q/Q growth rate, 4-quarter moving average) 4 3 2 1 0 -1 -2 UK -3 lowest growth rate among G7 highest growth rate among G7 -4 2012 2013 2014 2015 2016 2017
Conceptual framework Decision Maker Panel Basic Data Impact of Brexit 7
Classic “Stochastic Volatility” Uncertainty Model Style of Black Scholes (1973), Dixit and Pindyck (1994), Hassler (1996) and Abel & Eberly (1996). Actual model in Bloom (2009).
Classic uncertainty shock drop and rebound Source: Bloom (2009)
But Brexit is a “Bayesian” Uncertainty Shock (with an additional negative first moment shock) Style of Bernanke (1983)
Conceptual framework Decision Maker Panel Basic Data Impact of Brexit 11
Firms are recruited from a Nottingham call center
Quick monthly internet survey – e.g. sales question
Quick monthly internet survey – e.g. sales question
Quick monthly internet survey – e.g. sales question
Quick monthly internet survey – e.g. sales question
Data quality looks good – for example, uncertainty and forecast errors Note : Uncertainty defined as subjective uncertainty from the DMP 5-bin responses. Forecast errors defined as |forecast-actual| growth over the following 12 month period
DMP sample size
DMP broadly matches the pattern in the UK Business Register Percentage of employment Manufacturing Construction Wholesale & Retail Transport & Info. Accom. & Food Real Estate Prof. & Admin. Human Health DMP unweighted BEIS Business register Other services 0 10 20 30 40
The majority of DMP respondents are finance directors or senior managers Percentage of respondents 66 70 60 50 40 30 20 15 10 10 5 4 0 CEO CFO Finance Financial Other Director Controller/ Manager/ Executive Position of DMP respondents
These CEOs and CFOs are more negative on Brexit than the general population Source : Question “ What is your overall view of Brexit? ”. “DMP” is the Decision Maker Panel survey of CEOS and senior managers. “ nmg ” is a national voter survey
So check response in frame (all 10+ employee firms) is uncorrelated with local Brexit vote share Ever respond to a survey if in the sampling frame (1) (2) (3) (4) Leave vote share -0.002 -0.004 0.002 0.002 (0.015) (0.015) (0.015) (0.015) Log of employment 0.014*** 0.008*** 0.008*** (0.001) (0.002) (0.002) Log of sales 0.003 0.003 (0.002) (0.002) 0.003* Log of assets 0.003* (0.002) (0.002) Latest real sales growth 0.000 (0.000) Latest profit margin -0.000 (0.000) Observations 39,757 39,757 39,757 39,757 R-squared 0.076 0.079 0.079 0.079 Robust standard errors in parentheses
Data quality looks good – for example, comparing DMP to Company Accounts 14 10 DMP: log(Emp.) in 15Q4 DMP: log(Sales) in 2015 12 8 10 6 8 4 2 6 2 4 6 8 10 6 8 10 12 14 BvD: log(Sales) in 2015 BvD: log(Emp.) in 2015 .4 .4 .3 .3 .2 .2 .1 .1 0 0 0 5 10 15 0 5 10 15 20 ln(Num. of Emp.) ln(Sales) DMP: ln(Emp) in 2015Q4 DMP: ln(Sales) in 2015 BvD: ln(Emp) in 2015 BvD: ln(Turnover) in 2015
Data quality looks good – macro aggregates Percentage change on a year earlier, % Percentage change on a year earlier, % 4.5 10 DMP: Nominal Sales Growth DMP: Price inflation 4 ONS: Nominal Total Final ONS: CPI rate change over 12m 3.5 8 Expenditure 3 2.5 6 2 1.5 4 1 2 0.5 16Q3 16Q4 17Q1 17Q2 17Q3 17Q4 18Q1 16Q4 17Q1 17Q2 17Q3 17Q4 18Q1 18Q2
Decision Maker Panel Basic Data Impact of Brexit 25
Uncertainty over when (and if) Brexit will happen Question: “ When do you expect the UK to leave the EU? ”
Uncertainty over costs and revenues Question: “ How much has the result of the EU-referendum impacted the level of uncertainty affecting your business? ” Brexit one of top 3 source of uncertainty 50 Brexit largest current source of uncertainty 40 % respondents 30 20 10 0 Demand Labour Supply Regulation Customs availability chains
Uncertainty is also rising over time 60 Aug-Sep 16 Feb-Apr 17 Percentage of respondents 50 Aug-Oct 17 Feb-Apr 18 Aug-Oct 18 Nov-Jan 19 40 30 20 10 0 Not One of 2 or 3 top Top source important many sources of sources uncertainty Question: “ How much has the result of the EU-referendum impacted the level of uncertainty affecting your business overall? ”
Uncertainty jumped in particular in Sept 2018 Percentage of respondents who see Brexit as in top 3 current sources of uncertainty 60 Year 3 average 55 50 45 Average of first 2 years 40 35 30 Aug-Sep Feb-Apr Aug-Oct Feb-Apr Aug-Oct Nov-Jan 16 17 17 18 18 19
Interestingly do not see this in realized/news uncertainty measures (Brexit as a Bayesian shock?) Standard deviations from average since 1997 EU Referendum 7 6 5 4 Stock market volatility 3 Policy uncertainty index 2 1 0 -1 -2 1997 2001 2005 2009 2013 2017 ‘Stock market volatility’ measure is FTSE all-share three-month option implied volatility. ‘Policy uncertainty index’ is constructed based on newspaper articles regarding policy uncertainty in The FT, The Sunday Times, The Telegraph, The Daily Mail, The Daily Express, The Times, The Guardian, The Mirror, The Northern Echo, The Evening Standard, and The Sun.
Brexit uncertainty by firm size and region East Midlands North East 10-49 South East East of England Employment Northern Ireland 50-99 West Midlands South West Wales 100-249 North West Yorks and Humber London 250+ Scotland 0 10 20 30 40 50 60 70 80 90 100 0 20 40 60 Percentage of firms with Brexit in Percentage of firms with Brexit in top 3 uncertainty sources top 3 uncertainty sources Note : The question asked ‘How much has the result of the EU Note : The question asked ‘How much has the result of the EU referendum affected the level of uncertainty affecting your business?’ . referendum affected the level of uncertainty affecting your business?’ . 31 31
Decision Maker Panel Basic Data Impact of Brexit 32
Brexit was a shock – so we can also use a classic difference in differences estimation… Source: oddschecker.com, University of Stirling Management School and Centre on Constitutional Change
…by using Pre -Brexit (2016H1) EU exposure Shares of sales that are exports to EU Shares of costs that are imports from EU Percentage of panel members Percentage of panel members 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 0% >0-10% >10-20% >20-50% >50% 0% >0-10% >10-20% >20-50% >50% Shares of sales covered by EU regulations Shares of workforce who are EU migrants Percentage of panel members Percentage of panel members 60 40 50 30 40 30 20 20 10 10 0 0 0% 1-10% 11-50% >50% <1% 1-5% 6-10% 11-20% >20%
Firms’ pre -Brexit (2016H1) EU exposure predicts the Brexit uncertainty increase extremely well Note: Uncertainty measured as the response to the question ‘ How much has the result of the EU referendum affected the level of uncertainty affecting your business?’ where “Not important”= 1, “One of many sources”= 2, “Top 2 or 3 factors”= 3 and “Top factor”= 4.
Recommend
More recommend
