410 likes | 555 Views
COSTS OF FINANCIAL INSTABILITY - GDP, INVESTMENT AND CONSUMPTION. Course on Financial Instability at the Estonian Central Bank, 9-11 December 2009 – Lecture 10. E Philip Davis NIESR and Brunel University West London e_philip_davis@msn.com www.ephilipdavis.com
E N D
COSTS OF FINANCIAL INSTABILITY - GDP, INVESTMENT AND CONSUMPTION Course on Financial Instability at the Estonian Central Bank, 9-11 December 2009 – Lecture 10 E Philip Davis NIESR and Brunel University West London e_philip_davis@msn.com www.ephilipdavis.com groups.yahoo.com/group/financial_stability
PAPER 1:CORPORATE FINANCIAL STRUCTURE AND FINANCIAL STABILITYby E Philip Davis and Mark Stone Published in Journal of Financial Stability
Structure of paper • Introduction • Literature review • The data and corporate financial structure • Corporate financial structure and financial stability – descriptive analysis • Econometric analysis • Conclusions
Introduction • Dimensions of corporate financing • Depth – quantity of financing available • Breadth – variety of financing options (loans, bonds, shares, trade credit – and liquidity) • Measure implications of structure for fragility via: • Measuring real expenditure responses to banking and currency crises and variation therein linked to balance sheet indicators • Examining shifts in size and composition of corporate financing during a crisis, controlling for normal cyclical changes (shifts may link to supply-side, i.e. rationing of finance or demand-side, i.e. corporate balance sheet adjustment)
Literature review • General determinants of corporate financial structure • Exceptions to Modigliani-Miller – costs of bankruptcy versus tax benefits to debt • Asymmetric information and higher cost of external financing (issue how well financial system deals with agency costs) • Economic and financial development • Overall development of financial services important to growth - and not bias to bank or market financing? • Stages of development in financial systems – stock and bond markets come later than banks. Legal aspects important
Financial effects on the business cycle • Financial accelerator and effect of net worth (debt/equity ratio) on availability of external finance • Credit channel and specialness of bank finance (bank debt/total debt ratio) • Recent theories of financial crises • Introduction of banks and liquidity effects into most recent models • Collateral effects and micro and macroeconomic rigidities • Financial breadth (“multiple avenues”) and the impact of crises on expenditures • Empirical work on post crisis output contractions • Macroprudential indicators of corporate sector fragility
Descriptive analysis of crises, growth and investment • Standard set of banking and currency crises (Eichengreen and Bordo 2002) – 59 events for 29 countries, 17 in EMEs, 18 banking crises and four twin crises • Methodology is to measure effects as deviation of contribution to GDP from trend effect over 11 years; 5 before crisis year and 5 after • Larger impact of crises on EMEs GDP – mainly accounted for by domestic demand and corporate investment expenditures and inventory decumulation therein. Consumption relatively robust and external demand expands • Banking crises greater effect than currency crises
Indicators of financial structure and crises • Size of balance sheets per se is unrelated to expenditure response… • …but marked impact of corporate leverage on investment and inventory contraction • Falls in flows of external finance mainly in bank loans and liquidity falls while bond issues rise. Falls in external finance and bank lending larger for EMEs and in banking crises
Econometric analysis • Estimate specifications capturing normal cyclical developments before testing for additional effects of crises • First estimated corporate expenditure functions with crisis and balance sheet variables • Then estimated flow of funds/GDP determinants • Cross section weighted GLS panel with fixed effects and cross section weights (common factors world growth, world trade, share prices, bond yields)
Results suggest strong crisis effects on both corporate expenditure and corporate financing • Crisis effects smaller for currency crises than banking crises and for OECD than EME • Effects on investment aggravated by high debt/equity (financial accelerator) and low bank/debt (credit channel) • Flow of funds equations suggest fall in bank lending, equity issue (information spillovers from bank lending) but rise in bond issue (multiple avenues of intermediation)
Bond issuance function Dependent variable: difference of bond issuance/GDP
Conclusion • Probed impact of crises on corporate financing and expenditures in OECD and EME countries • Marked differences in balance sheet size and composition as well as flows between OECD and EME countries • Investment and inventory contractions key components of GDP decline in wake of crises, notably in EMEs, underlining importance of better understanding of dynamics of corporate behaviour during crises • Role of debt equity ratio in determining impact as well as bank lending/debt • Role of bond market as shock absorber in OECD
Evidence strengthens the case for financial sector reforms, notably in EMEs, given role of external finance. • Bank lending and trade credit warrant particular focus in EMEs • Development of bond markets also warranted • IMF surveillance needs to focus on corporate sector performance in macroprudential surveillance and not merely banking sector – and also in Article IV in assessing monetary policy transmission during a crisis • Encourage countries to produce flows of funds including back data • Further research in this overall area desirable
PAPER 2: COSTS OF FINANCIAL INSTABILITY, HOUSEHOLD-SECTOR BALANCE SHEETS AND CONSUMPTION Ray Barrell, E Philip Davis and Olga Pomerantz Published in Journal of Financial Stability
Introduction • Literature on costs of financial instability tends to focus on fiscal costs and the impact on GDP of banking crises. • We analyse the effect of a banking or currency crisis on consumption. • We show consumption plays an important role in the macroeconomic adjustment following a financial crisis • Effect of a crisis is aggravated by high leverage, also greater in a small open economy
Falling house prices shown to be part of the transmission process of financial instability • High nominal interest rates are an indicator of sharp declines in consumption • Results imply that a banking crisis taking place now could have a greater incidence than in the past, especially if macroeconomic policy is unable to respond, as for a small country in EMU. • Also highly relevant as Basel II encourages consumer lending via lower risk weights
1 Measuring costs of banking crises • More studies on prediction than costs • Components of costs • Losses by stakeholders in failing banks • Losses by borrowers • General recession hitting consumption and investment (but crises also occur in recessions) • Fiscal resolution costs, which also affect others • Issue of timing of crisis, linked to definition
Fiscal costs ca. 12% of GDP (larger in EMEs), depending on role of banks, scope of liquidity support • Output losses usually growth or (Hoggarth-Sapporta) output level difference from trend or recession in comparable countries • Stylised facts crises last longer in OECD countries and output costs can be greater • Also twin crises costlier than banking crises • Davis-Stone components of GDP analysis – investment bears the brunt (-3.1% of GDP) compared to consumption (-1.4% of GDP) • Econometric approach – dummies in investment function – that we follow here
2 The impact of banking and currency crises on consumption • Probe the nature and determinants of reaction of consumption to financial instability over and above impact of crisis on real personal disposable income and net financial wealth • Test for dummies in standard consumption function, also with focus on financial liberalisation and resulting rise in leverage • Standard Caprio-Klingebiel definitions of crises: • Currency crises entail forced change in parity, abandonment of a pegged exchange rate, or an international rescue. • Banking crisis involve bank runs, widespread bank failures and the suspension of convertibility of deposits into currency, or significant banking sector problems that result in the erosion of most or all of banking system capital.
Consumption specification • Function of human and non human wealth, where human proxied by its return (RPDI) • Log transform to allow for growth • Comparable work Ludvigsen-Steindel, Davis-Palumbo, Barrell- Davis • Panel GLS estimation, error correction
Points to note • Crises take time to impact and effects build • Effects largest in small open economies and since 1990 • Role for leverage in augmenting eventual effects • Higher gearing, higher effect of liquidity constraints…but high gearing in liberalised systems with less liquidity constraints • Mean D/Y 0.65 to 1989, 0.84 from 1990 • In first year crisis effect ameliorated by high D/Y, but in second makes substantially worse
Conclusions • Literature focuses largely on fiscal costs and impact of crises on GDP • Seen that consumption key element of macroeconomic adjustment • Effect is worsened by high leverage, despite the benefit of liberalisation • Also more serious in Small Open Economies – especially if macro policy cannot adjust to offset • Need for macroprudential focus on household leverage
References • Barrell, R., Davis, E.P., and Pomerantz, O., (2006) "Costs of Financial Instability, Household-Sector Balance Sheets and Consumption", Journal of Financial Stability, 2, 194-216 • Davis E P and Stone M (2004), "Corporate Financial Structure and Financial Stability", Journal of Financial Stability, 1, 65-81