Lack of international risk management in BREXIT?

Abstract

BREXIT is a historical step for the UK and the EU27 which could bring a strong Pound deprecation, an increase in risk premiums for British bonds and a transitory rise of financial market volatility plus a long-term reduction of economic growth in the UK; but also risks for the EU27. New BREXIT aspects are considered in an enhanced Branson model. Macroprudential supervision is a crucial policy challenge for EU28 in the context of BREXIT and the European Systemic Risk Board thus should have a critical role in 2019 and the following years. The ESRB should timely analyze the potential risk of BREXIT and consider adequate policy options to reduce or eliminate risks. Contract continuity as well as cooperation in prudential supervision between the EU27 and the UK stand for BREXIT-related problems that could create financial market stability – as is the BREXIT-induced UK deregulation pressure. EU prudential supervision post-BREXIT faces problems since a very large part of EU27 wholesale banking markets in the UK and thus not regulated by the EU after March 29, 2019. The EU Commission’s competence for EU trade policy as well as international investment treaties gives the EU the opportunity to offer the UK not only a – limited – Free Trade Agreement but an international investment treaty as well, including options for global cooperation. Several policy innovations are proposed which could help to limit risk associated with instability.

Appendices

Appendix 1: A Simple BREXIT Macro Analysis of the Goods Market Equilibrium in the Context of an Asymmetric FDI

The impact of BREXIT on the UK is crucial, but there is also the question of to what extent the EU27 or the Eurozone will be affected by the UK’s leaving of the EU. Consider a simple macro model (without government expenditures, but with cumulated FDI inflows) where one can write the goods market equilibrium condition for the medium term as (with positive parameters c, x, j – the three parameters are in the range of 0,1 - and q* denoting the real exchange rate eP*/P:=q* and τ is the income tax rate; Y is real gross domestic product, I is real investment, exports X are proportionate to foreign real gross national income (* for foreign variable, α is the share of country-1 investors in the foreign capital stock of country 2, ß* is the share of profits in foreign gross domestic product) Z*:=Y*(1-αß*) and also a positive function of q* (by assumption with an elasticity of one); hence X = xq*Z* Import volume J is assumed to be proportionate to disposable national income (J = j(…)Y) and to be a negative function of q* (with an elasticity of imports with respect to q* of −1), real imports expressed in domestic goods units are q*jZ so that we can write (with consumption being proportionate to disposable national income Z:= (Y + αß*q*Y*)) in a stochastic context with a white noise error term ε:

$$ \mathrm{Y}=\mathrm{c}\left(1-\uptau \right)\left(\mathrm{Y}+\alpha {\beta}^{\ast }{\mathrm{q}}^{\ast }{\mathrm{Y}}^{\ast}\right)+\mathrm{I}+{\mathrm{xY}}^{\ast}\left(1-\alpha {\beta}^{\ast}\right){\mathrm{q}}^{\ast }-\mathrm{j}\left(1-\uptau \right)\left(\mathrm{Y}+\alpha {\beta}^{\ast }{\mathrm{q}}^{\ast }{\mathrm{Y}}^{\ast}\right)+\upvarepsilon $$

(10)

$$ \mathrm{Hence}\ \mathrm{Y}=\mathrm{I}/{\mathrm{s}}^{'}+\left[\right({\mathrm{xq}}^{\ast }/{\mathrm{s}}^{'}+\left(\left(\mathrm{c}-\mathrm{j}\right)/{\mathrm{s}}^{'}\right)\left(1-\uptau \right)\alpha {\beta}^{\ast }{\mathrm{q}}^{\ast}\Big]{\mathrm{Y}}^{\ast }+\upvarepsilon; \mathrm{where}\ {\mathrm{s}}^{'}:= 1-\mathrm{c}\left(1-\uptau \right)+\mathrm{j} $$

(11)

The expectation value E(Y) therefore is

$$ \mathrm{E}\left(\mathrm{Y}\right)=\left(1/{\mathrm{s}}^{'}\right)\mathrm{E}\left(\mathrm{I}\right)+\left[{\mathrm{xq}}^{\ast }/{\mathrm{s}}^{'}+\left(\left(\mathrm{c}-\mathrm{j}\right)/{\mathrm{s}}^{'}\right)\left(1-\uptau \right)\alpha {\beta}^{\ast }{\mathrm{q}}^{\ast}\right]\mathrm{E}\left({\mathrm{Y}}^{\ast}\right) $$

(12)

Moreover, with s”:= 1/s’– we have for the variance

$$ \mathrm{V}\left(\mathrm{Y}\right)={\mathrm{s}}^{"2}\mathrm{V}\left(\mathrm{I}\right)+{\left[\left({\mathrm{xq}}^{\ast }+\right)\left(\mathrm{c}-\mathrm{j}\right)\left(1-\uptau \right)\alpha {\beta}^{\ast }{\mathrm{q}}^{\ast}\right]}^2{\mathrm{s}}^{"2}\mathrm{V}\left({\mathrm{Y}}^{\ast}\right)+2{\left[\left({\mathrm{xq}}^{\ast }+\left(\mathrm{c}-\mathrm{j}\right)\right)\left(1-\uptau \right)\alpha {\beta}^{\ast }{\mathrm{q}}^{\ast}\right]}^2{\mathrm{s}}^{"4}\operatorname{cov}\left(\mathrm{I},{\mathrm{Y}}^{\ast}\right); $$

(13)

The country’s investment will be proportionate to the foreign country’s GDP. Obviously the share of cumulated ownership abroad (α) contributes to a higher variance of Y. A necessary and sufficient condition for this is a positive covariance.

Here it is assumed that the higher the export-GDP share, the higher should be the covariance cov(I,Y*); from a Eurozone country perspective – with UK being the foreign country – the output variance V(Y) is raised in a double way through the UK’s BREXIT. The variance of the UK’s output will increase which, in turn, will raise the variance of the Eurozone country’s output. Moreover, the covariance cov(I,Y*) could increase as investment in the Eurozone country will be correlated positively with Y*. If the Eurozone country is itself a strong producer of capital equipment, the implication should be that the rate of return of equipment producers is reduced so that stock market values of that sector will fall. There is a caveat to this view since import tariffs in some sectors of EU countries will indirectly stimulate British tariff jumping investment in Eurozone countries. This in turn should stimulate production of capital equipment in major capital equipment producer countries. If exports to the UK are replaced in the medium term and long term by outward foreign direct investment and UK production, respectively, the covariance cov(I,Y*) should fall. Hence the output variance in Eurozone countries should increase temporarily. To the extent that BREXIT has negative output effects on the EU27 – here the link should be (following standard QUEST results from the EU Commission’s macro model) roughly that 6% income reduction in the UK will bring about 1% GDP reduction in the EU27 – there will be a negative repercussion effect on UK output. Part of negative output effects in EU27 countries could be linked to slightly higher financing and hedging costs that are associated with a relocation of banking activities from the “City of London”, usually considered to represent big economies of scale effects, to EU27 countries. If there is a hard BREXIT, in the sense that there is not EU-UK treaty on BREXIT, there could be large financial shocks in the UK and the EU27 countries whose main banking wholesale market is in London.

Appendix 2: Implied Economic Effects of BREXIT – Forecast Revisions

Fig. 10

Impact of revised office for budget responsibility forecasts for UK real GDP. Source: Welfens (2017), The True Cost of BREXIT for the UK: A Research Note, EIIW Discussion Paper No. 234, www.eiiw.eu (Base year 2016 = 100)

Fig. 11

Impact of revised OBR forecasts for UK inflation. Source: Welfens/Hanrahan (2018), based on OBR, Historical Official Forecast Database, http://budgetresponsibility.org.uk/data/

Fig. 12

Impact of revised OECD forecasts for UK inflation. Source: Welfens/Hanrahan (2018), based on OECD Economic Outlook, 2017 Issue 2 and 2015 Issue 2

Appendix 3: FDI impediments in OECD countries

Table 3 World currency composition of official foreign exchange reserves in US dollars, millions

Table 4 World currency composition of official foreign exchange reserves (Share in Total Foreign Exchange Reserves in Percentage)

Appendix 4: Household Net Worth in the Great Depression and the Great Recession

Fig. 13

Comparing US output loss in the great depression and during the great recession (2008 onwards). Source: Council of Economic Advisers (2017). Economic Report of the President. p. 29

Appendix 5: Possible Future UK-EU Relationship

Fig. 14

Possible Future Relationship Scenarios (slide presented by EU Chief Negotiator Michel Barnier to European Commission. Source: European Commission (2017). https://ec.europa.eu/commission/sites/beta-political/files/slide_presented_by_barnier_at_euco_15-12-2017.pdf

Appendix 6: The EU macroprudential policy framework (Source: European Parliament (2017), IPOL/EGOV)

Fig. 15

ESRB governance

Fig. 16

CRD IV/CRR macroprudential toolkit

Appendix 7 OECD FDI regulatory restrictiveness analysis

Fig. 17

Linear relationship between total with Banking&Insurance and total with financial services

The relationship displayed in the scatterplot above is not monotonic, thus for a deeper analysis either a transformation or another type of test entirely would be appropriate. However, most of the data is 0, therefore the suggestion would be change to another test. The upper two data points may have to be removed as outliers for Norway and Hungary. Even though using a Spearman’s correlation test on the current dataset might not lead to a valid result, the output appears below:

A Spearman’s correlation was run to assess the relationship between Total with Banking&Insurance, and Total with Financial Services using a small sample of 36 countries. There was a weak positive correlation between Total with Banking&Insurance and Total with Financial Services, which was not statistically significant at the level of .05, rs = .0115, p = .9478.

Fig. 18

Linear relationship between total sectors and Banking&Insurance

Fig. 19

Linear relationship between total sectors and financial services

We have a positive significant correlation between Total and Financial Services (at the significance level of .05).

Appendix 8: Banking and Insurance in Europe

Many insurance companies and financial conglomerates, respectively (in Europe there were 83 conglomerates in 2016, up from 75 in 2009) could be exposed to BREXIT risk and increased risk premiums on Pound-denominated bonds in the context of BREXIT. The knowledge about systemic risk is rather limited; insurance companies could be exposed to spillovers from banks on the one hand, on the other hand insurance companies could themselves be a source of spillovers for other financial sector institutions.

The ESRB has published a report on insurance companies in 2017 (ESRB, 2017, Recovery and resolution for the EU insurance sector: a macroprudential perspective, August 2017, Report by the ATC Expert Group on Insurance). One may quote some critical insights here as a quote:

“Within Europe and North America respectively, there could be large spillovers a between different sectors, including insurers. In Asia, non-life insurers and reinsurers seem to be highly interconnected with other sectors in the region. In terms of spillovers across the regions, Europe and North America appear to be the most interconnected, with insurers (in particular life insurers) from Europe having a high potential to transmit spillovers to the American financial market (IMF2016a)....A separate analysis for Europe indicated that, before the global financial crisis, insurers were recipients of spillovers from other sectors although, more recently, they seem to have become a source of spillovers (IMF2016a). Insurers may pose systemic risks arising from their funding and investment activities. Collectively, insurers are among the largest investors in financial assets in the EU. They can contribute to systemic risks through various channels, which include taking up more risks, increasing commonality in asset composition within the financial sector, leading to increased exposure to common shocks (“tsunami risk”), or increasing procyclicality in their investment behaviour. For instance, analysis by the Bank of England concludes that the systemic risk associated with activities of the UK insurance sector that propagate or amplify shocks to financial counterparties or markets may be the greatest source of systemic risk from insurers for the UK (French et al. 2015)...

Disruption to systemically important financial counterparties can occur if these institutions no longer have access to funding from EU insurers. Insurers hold large amounts of debt securities and shares issued by banks and other financial institutions in the EU. From the perspective of banks’ balance sheets, these accounted for 4% of total bank funding in the euro area in 2014 (ESRB 2015), while, on average, around 13% of debt issues by euro area banks is held by insurers domiciled in the euro area and Sweden...This figure is even higher in some EU Member States, e.g. 28% in Belgium, Greece and Slovakia and 37% in France. The ECB has emphasised that contagion risks from ownership links to banks and other financial institutions are among the most significant risks (ECB 2008).”

Appendix 9: Degree of EU Liberalization in CETA

Fig. 20

Degree of EU liberalization in CETA. Source: Magntorn, J.; Winters, A.L. (2018)