Management of fiscal risks: macroeconomic scenarios, contingent liabilities

Transcription

1 EUROPEAN COMMISSION DIRECTORATE GENERAL ECONOMIC AND FINANCIAL AFFAIRS Managemen of fiscal risks: macroeconomic scenarios, coningen liabiliies Lucio Pench Head of Uni Public finances in he euro area and he EU IMF Paris, Ocober 2008

2 Misaken assumpions on he exernal environmen are in good par o blame The exernal environmen explains a relaively imporan par of he forecas error (up o 60% of he error made in forecasing GDP or inflaion a EU level); The forecas error explained by he exernal environmen increases when he forecas horizon lenghens and less informaion is available; The inernaional environmen has a greaer responsibiliy in misakes in EU GDP and inflaion han assumpions on ineres raes, exchange raes and oil prices (Keereman, 2003). Forecas errors for GDP: Tes for unbiasedness Curren-year forecas Bias (unbiasedness es) α (ME) Signif. α=0 EU euro area Year-ahead forecas EU euro area Source Melander e al,

3 Some issues for discussion I. Macroeconomic scenarios: he key is he medium erm; II. Coningen liabiliies: unknown unknowns. 3

4 Making predicions is difficul, especially abou he fuure Economic forecass are key o budgeary projecions. Ex-pos, lower/higher-han-projeced growh affecs he fiscal sance. Commission s forecass ouperform naïve ones, bu sill forecas for GDP growh has, on average, proven o be 0.5 pp. oo high/low even for he curren year (Melander e al., 2007). Overall, no marked improvemens have been recorded in he qualiy of he forecass over ime. Forecas rack record of IMF, OECD, Consensus and Commission is broadly comparable, wih he iming of he forecas playing a subsanial role. Forecas errors for GDP Mean Error Mean Absolue Error Roo Mean Squared Error Sample curren year year ahead curren year year ahead curren year year ahead EU 69/ euro area 98/ Source Melander e al, 2007

5 Inernaional forecass a leas no as biased as some official growh forecass Source Jonung and Larch,

6 Biased forecass of real GDP affec poenial oupu esimaes and hence he fiscal sance Upward bias in real GDP growh ranslaes in opimisic assessmen of poenial growh; If poenial GDP is overesimaed, a fiscal policy ha appeared pruden ex ane migh resul as expansionary ex pos. 6 Source Jonung and Larch, 2006

7 Independen forecass as a way o improve accuracy Mos EU Member Saes rely on in-house macroeconomic forecass for heir budgeary plans. The few excepions are Ausria, Belgium and he Neherlands. 7 Source Jonung and Larch, 2006

8 Does long-erm growh maer for susainabiliy? Impac on he susainabiliy indicaor of changes in macroeconomic assumpions is surprisingly limied; Wha is more imporan is he saring poin: budgeary consolidaion on he medium erm can very efficienly limi he public finance susainabiliy challenge over he long-erm. Impac on he S2 indicaor (main scenario) Higher life expecancy Higher Higher Higher employmen labour employmen Toal Impac on produciviy of older If due o If due o worker an an Pensions Healh Longerm increase decrease care in labour in he care supply NAIRU Higher ineres rae Toal impac IBP of which LTC EU EU Sandard deviaion

9 Governmen expendiure reacs o poenial oupu dynamic in he medium erm On average, long-run elasiciy of governmen expendiure wih respec o GDP in EU-15 is slighly below uni, meaning he expendiure is linked o poenial oupu by roughly a one-o-one relaionship; I is significanly higher han uniy in caching-up counries, in fas-ageing counries, in low deb counries and in counries wih weak numerical rules for he conrol of governmen spending; On average, governmen expendiure is adjused o poenial oupu in abou 3 years; Anglo-Saxon and Nordic counries exhibiing general higher speed of adjusmen han Souhern European counries (Arpaia and Turrini, 2008). 9

10 A axonomy of governmen liabiliies according o he degree of cerainy known knowns Explici (governmen obligaions have legal basis) Implici (governmen obligaions do no have legal basis and arise as a consequence of expecaions creaed by pas pracice or pressures by ineres groups) Non-coningen liabiliies (he exisence of governmen Coningen liabiliies (he exisence of obligaions obligaions does no depend upon paricular evens) depends upon he realizaion of paricular evens) I Governmen deb Well known II Governmen individual guaranees on he deb issued by public and privae eniies Governmen expendiures as saed in budge law Governmen umbrella guaranees (e.g., on household morgages, ) Provisions (e.g., accrued-o-dae pension righs from unfunded schemes) Governmen insurance schemes (on bank deposis, on reurns from privae pension funds, ) III IV Fuure welfare paymens (pension paymens Bail ou of defauling public secor or privae relaed wih pension righs which have no maured ye, eniies (public corporaions, banks or oher privae fuure healh care paymens, ) financial insiuions, pension and social securiy funds, ) Fuure governmen expendiures relaed o Disaser relief recurren operaions (e.g., capial sock refurbishmen, ) Environmenal damage Miliary financing 10 unknown unknowns (known only ex pos)

11 Boundaries of governmens affec wha we know abou fiscal risks In ESA, governmen-conrolled unis are classified in he corporae secor or in governmen depending wheher sales cover more or less han 50% of coss. This crierion makes sense in compilaion of GDP, when here is a need o decide wheher oupu should be valued according o prices or according o coss, i makes less sense in fiscal surveillance. A public enerprise whose sales cover only 51% of is coss is no viable wihou coninuous governmen suppor: ulimaely hese coss end up in he governmen defici. More sringen rules on he secoral classificaion of public enerprises would improve risk monioring. 11

12 Y Y 1 1 = Y + Y 1 1 * * y + Y Sock-flow adjusmen capures hidden deficis linked o wider public secor operaions (% of GDP) average COM SP COM SP COM SP SP SP Gross deb raio Change in he raio Conribuions 2 : Primary balance Snow-ball effec Sock-flow adjusmen Of which: Cash/accruals diff n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Acc. financial asses n.a n.a n.a Privaisaion n.a. 0.5 n.a. 0.4 n.a Val. effec & residual n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Noes: 1 End of period. 2 The change in he gross deb raio can be decomposed as follows: Deb dynamics: Porugal D Y D Y 1 1 PD = Y D + Y 1 1 i * 1 + y y SF + Y where is a ime subscrip; D, PD, Y and SF are he sock of governmen deb, he primary defici, nominal GDP and he sock-flow adjusmen respecively, and i and y represen he average cos of deb and nominal GDP growh (in he able, he laer is decomposed ino he growh effec, capuring real GDP growh, and he inflaion effec, measured by he GDP deflaor). The erm in parenheses represens he "snow-ball" effec. The sock-flow adjusmen includes differences in cash and accrual accouning, accumulaion of financial asses and valuaion and oher residual effecs. Source: Sabiliy programme (SP); Commission services auumn 2007 economic forecass (COM); Commission services calculaions 12

13 Risks go beyond he public secor: fiscal cos of some pas banking crises Counry Period Fiscal Cos % of GDP Explici Inervenion/Resoluion Policy Tools Guaranee Liquidiy suppor Forbearance > 75% saeowned Deposi o DMB o NBFIs Freezes A B C Repeaed Recaps Public AMC Public Deb Relief Programme Ausralia no no no no no no yes no no no no Finland yes no yes - no no yes no no yes no France no no no no no no yes no no yes no Hungary no yes yes - yes no no yes yes no no Japan yes no yes - no no yes yes yes no no New Zealand no no yes - no no no no no no no Norway yes no yes - no no yes no no no no Poland no yes yes - no no yes yes no no no Slovenia yes yes no no yes yes no yes no yes no Spain no no yes - no no yes no no no no Sweden yes no no no yes no no no no yes no Turkey no no no no yes no no no no no no Turkey yes - no no yes no no yes no no no Unied Saes no no no no yes yes yes yes no no no Source: Honohan, Klingebiel,

14 A few rough numbers on he fiscal cos of he curren financial crisis % of GDP Recapialisaion Effecive Funds sill o be used Special veichles Guar. liquidiy faciliy a CB Guaranee on shor erm borrowing by banks Fees and dividends Asses exchange/purchase and loans Sae Guaranee on Deposis Budgeed Poenial Poenial Budgeed Poenial Toal ne of fees BE 3.5 no fig Unlimied BG CZ - - DK 2.4 Possible Unlimied Unlimied 2.4 DE no fig. Unlimied 20.8 EE Unlimied EL Unlimied 11.5 ES FR IE Unlimied IT no fig. no fig. 2.5 CB + no fig. Gov Unlimied 2.5 CY - - LV no fig. LT LU HU MT - - NL AT Unlimied 42.1 PL no fig. no fig. PT RO SI Unlimied SK Unlimied FI no fig. no fig. SE UK EU > > > EU > > >

15 A complee surveillance should look a ne worh: balance shees BE BG CZ DK DE EE IE EL ES FR IT CY LV LT LU HU MT NL AT PL PT RO SI FI SE UK Source: Commission Services 15