Real Sector Division IMF Statistics Department Adding Creditor

  • Slides: 15
Download presentation
Real Sector Division IMF Statistics Department Adding Creditor Perspective Information to National Accounts Marshall

Real Sector Division IMF Statistics Department Adding Creditor Perspective Information to National Accounts Marshall Reinsdorf (IMF) and Kyle Hood (Bureau of Economic Analysis) Society for Economic Measurement Conference Paris, July 23, 2015 The views expressed herein are those of the authors and should not necessarily be attributed to the IMF, its Executive Board, or its management, or to Bureau of Economic Analysis

Real Sector Division IMF Statistics Department Bad Debt in National Accounts § In the

Real Sector Division IMF Statistics Department Bad Debt in National Accounts § In the current accounts, disposable income and saving are calculated as though all amounts due were paid • • • Commodities delivered but not paid for are defined as part of production, and income from production Defaulting is not a transaction The picture of debtors’ finances must include all payables accrued § Reductions in loan balance by agreement are capital transfers § to the borrower, but loan write-offs due to defaults fall under “Other changes in volume of assets” But the SNA does recommend a balance sheet memo item showing the market value of loans or the nominal value net of provisions for expected losses

Real Sector Division IMF Statistics Department Expected losses on loans to risky borrowers §

Real Sector Division IMF Statistics Department Expected losses on loans to risky borrowers § In business accounting, lenders deduct provisions for § § § expected default losses when calculating their income But to measure borrowers’ saving, the entire amount of interest charges and debt that they incur has to be counted, even if the probability of default > 0 To avoid inconsistencies in system of accounts, approach that is appropriate for borrowers is also applied to lenders Expected default losses are not considered when calculating the disposable income of lenders

Real Sector Division IMF Statistics Department The Lending Business § Default losses are a

Real Sector Division IMF Statistics Department The Lending Business § Default losses are a normal part of the lending business, § § § much like claims are part of the insurance business Besides a “return-to-risk-bearing” premium to generate the return required on a risky investment, interest rates to risky borrowers include a default premium lender can substitute interest for missing repayments of principal With no provision for expected losses, income from loans to risky borrowers is overstated in national accounts During times of rising losses, such as the financial crisis, the income of banks as measured in national accounts may show implausible growth

Real Sector Division IMF Statistics Department Measuring the output of lenders § Implicitly-priced financial

Real Sector Division IMF Statistics Department Measuring the output of lenders § Implicitly-priced financial intermediation services are § § measured by FISIM reclassifies some interest from loans as an implicit payment for services, leaving a “pure interest” residual FISIM consumed by borrowers is measured by the spread between the loan rate and the reference rate But some of this interest is actually needed to cover expected losses of principal due to default The default margin can be netted out of the loan rate used to calculate FISIM without changing the measure of lenders’ income by letting their “pure interest” get bigger

Real Sector Division IMF Statistics Department Example of average loss rate and FISIM q

Real Sector Division IMF Statistics Department Example of average loss rate and FISIM q Default margins are modest for some types of loans, but large for others q For US commercial banks in 2013, the average charge-off rate on loans in general was 0. 7 percent, while the average effective interest rate (which excludes uncollectable interest) was 5. 9 percent. q But some types of loans, like credit cards, have much higher loss rates Credit Card Lending in the UK, averages for 1994 -2007 User cost margin for FISIM, conventional approach 7 percent Write-off rate 3. 5 percent User cost margin for FISIM, corrected 3. 5 percent

Real Sector Division IMF Statistics Department Alternative measure of income of lenders § A

Real Sector Division IMF Statistics Department Alternative measure of income of lenders § A measure of income of lenders that is appropriate for many § purposes nets out a provision for expected default losses To exclude the default premium from both “pure interest” and FISIM when looking at the lender but not when looking at the borrower, the system of accounts can include an adjustment item equal to the provision for expected losses • “Taxes on products” is an example of such an adjustment item § But we prefer a simpler method § To avoid disrupting the core accounts, alternative measures of disposable income and saving of the financial corporations that make loans can be shown as supplementary information

Real Sector Division IMF Statistics Department Imputed capital transfers to borrowers § Lenders to

Real Sector Division IMF Statistics Department Imputed capital transfers to borrowers § Lenders to risky borrowers do not expect to receive back all of the principal; they just expect not to lose more than the interest generated by the default premium § An alternative way to look at borrowers would be to record funds advanced to risky borrowers in the expectation of not being paid back as imputed capital transfers § Role of amounts that are advanced and not paid back as a source of funding for spending would be illuminated § This alternative approach to borrowers can also be applied to creditors and will give the same measure of their net lending as deducting expected losses from their saving

Real Sector Division IMF Statistics Department Examples § Our examples for the US and

Real Sector Division IMF Statistics Department Examples § Our examples for the US and France show bigger effects § § of the alternative treatments of expected losses in the case of the US, implying that this information we are recommending is important for international comparisons To estimate expected losses in the US we apply an adaptive expectations with an adjustment speed of 0. 3 per year to data on charge-off rates by type of loan The calculations for France just used simple averages over the time period

Real Sector Division IMF Statistics Department Saving less Net Capital Transfers Paid, Financial Corporations

Real Sector Division IMF Statistics Department Saving less Net Capital Transfers Paid, Financial Corporations in the US (billions of US dollars) Saving less net capital transfers paid with no correction for expected losses Corrected for expected default losses of commercial banks Corrected for expected losses of financial corporations on commercial bank loans and home mortgages Memo: Expected default losses, commercial banks loans and home mortgages Net lending with no correction for expected credit losses 2007 2008 2009 2010 2011 -48. 5 -68. 9 293. 9 249. 9 182. 1 -86. 0 -116. 9 219. 8 137. 3 67. 9 -103. 0 -161. 1 132. 8 49. 8 -13. 5 54. 5 92. 2 161. 1 200. 1 195. 6 -94. 6 -91. 2 304. 4 263. 8 183. 3

Real Sector Division IMF Statistics Department Saving less Net Capital Transfers Paid, Households and

Real Sector Division IMF Statistics Department Saving less Net Capital Transfers Paid, Households and NPISH in the US (percent of disposable personal income) No correction for expected losses Corrected for expected default losses of commercial banks Corrected for expected losses of financial corporations on commercial bank loans and home mortgages 2007 2008 2009 2010 2011 2. 9 4. 9 6. 4 5. 8 6. 0 3. 1 5. 1 6. 9 6. 5 6. 8 3. 3 5. 5 7. 7 7. 3 7. 5 24. 5 28. 9 51. 7 82. 2 90. 2 42. 1 74. 5 141. 2 172. 7 174. 7 Memo: Imputed capital transfers received (billions of US $) from commercial banks from financial corporations on commercial bank loans and home mortgages

Real Sector Division IMF Statistics Department Corrected income from interest and FISIM of depository

Real Sector Division IMF Statistics Department Corrected income from interest and FISIM of depository corporations in France 2008 2009 2010 2011 2012 Correction for expected losses 324 106 430 7 202 120 322 10 151 127 278 8 180 123 303 10 171 126 297 13 Corrected total income from SNA interest + FISIM 424 312 270 293 284 Corrected FISIM 100 119 114 113 Expected default losses, as % of borrower FISIM 1. 6 6. 6 3. 1 8. 3 2. 9 6. 3 3. 3 4. 4 8. 1 10. 3 Expected default losses, as % of loans outstanding 0. 1 0. 2 “Pure interest” received FISIM, no correction for expected default losses Total income from SNA interest + sales of FISIM Memo items: Expected default losses, as % of total interest received 0. 2

Real Sector Division IMF Statistics Department Imputed capital transfers received, nonfinancial corporations in France

Real Sector Division IMF Statistics Department Imputed capital transfers received, nonfinancial corporations in France 2008 2009 2010 2011 2012 115 38 153 173 87 52 139 165 68 57 125 191 80 55 135 191 72 53 125 185 4 6 4 5 7 171 195 196 192 Imputed capital transfers, as % of loans outstanding 2. 6 0. 2 4. 3 0. 3 3. 2 0. 2 3. 7 0. 3 5. 6 0. 3 Imputed capital transfers, as % of disposable income 2. 6 4. 3 3. 2 3. 7 5. 6 “Pure interest” paid FISIM, no correction for expected default losses Pure interest paid + FISIM consumed (not corrected) Gross disposable income, national accounts Imputed capital transfers received Gross disposable income + imputed capital transfers Memo items: Imputed capital transfers, as % of total interest paid

Real Sector Division IMF Statistics Department Imputed capital transfers received, households in France 2008

Real Sector Division IMF Statistics Department Imputed capital transfers received, households in France 2008 “Pure interest” paid FISIM, no correction for expected default losses Pure interest paid + FISIM consumed (not corrected) Gross disposable income, national accounts Imputed capital transfers received Gross disposable income + imputed capital transfers 2009 2010 2011 2012 65 57 72 67 54 2 11 17 13 17 67 68 89 80 71 1579 1592 1663 1666 1681 2 4 4 4 5 1581 1596 1667 1670 1686 Memo items: Imputed capital transfers, as % of loans outstanding 3. 0 0. 2 5. 9 0. 4 4. 5 0. 4 5. 0 0. 4 7. 0 0. 5 Imputed capital transfers, as % of disposable income 0. 1 0. 3 0. 2 0. 3 Imputed capital transfers, as % of total interest paid

Real Sector Division IMF Statistics Department Conclusion § Risky borrowers pay higher interest rates

Real Sector Division IMF Statistics Department Conclusion § Risky borrowers pay higher interest rates so the lender § § § can substitute interest for expected losses of principal Ignoring losses from defaults gives a reasonable picture of the debtor’s situation, but is not suitable for creditors Disposable income, saving and “net lending” of financial institutions that make loans to risky borrowers are overstated Supplementary information on alternative measures that take account of expected default losses would allow the national accounts to present a more complete and meaningful picture