OECD FORECASTS DURING AND AFTER THE FINANCIAL CRISIS A

Oecd Forecasts During And After The Financial Crisis A-PDF Download

  • Date:13 Sep 2020
  • Views:0
  • Downloads:0
  • Pages:15
  • Size:316.60 KB

Share Pdf : Oecd Forecasts During And After The Financial Crisis A

Download and Preview : Oecd Forecasts During And After The Financial Crisis A


Report CopyRight/DMCA Form For : Oecd Forecasts During And After The Financial Crisis A


Transcription:

This Policy Note is published on the responsibility of the Secretary General of the OECD The opinions. expressed and arguments employed herein do not necessarily reflect the official views of the Organisation. or of the governments of its member countries, This document and any map included herein are without prejudice to the status of or sovereignty over any territory to the. delimitation of international frontiers and boundaries and to the name of any territory city or area. The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities The use of such data by the. OECD is without prejudice to the status of the Golan Heights East Jerusalem and Israeli settlements in the West Bank under the terms. of international law, You can copy download or print OECD content for your own use and you can include excerpts from OECD publications databases and. multimedia products in your own documents presentations blogs websites and teaching materials provided that suitable. acknowledgment of OECD as source and copyright owner is given All requests for public or commercial use and translation rights. should be submitted to rights oecd org Requests for permission to photocopy portions of this material for public or commercial use. shall be addressed directly to the Copyright Clearance Center CCC at info copyright com or the Centre fran ais d exploitation du droit. de copie CFC at contact cfcopies com, OECD FORECASTS DURING AND AFTER THE FINANCIAL CRISIS A POST MORTEM. Main Findings, GDP growth was overestimated on average across 2007 12 reflecting not only errors at the height of the. financial crisis but also errors in the subsequent recovery. Forecast errors were larger in countries that are more open to external developments and hence exposed to. shocks from other economies, Larger forecast errors over 2007 12 have occurred in countries with more stringent pre crisis labour and.
product market regulations, Growth in the recovery has been weaker relative to predictions in countries in which banks had low capital. ratios pre crisis, Stronger projected fiscal consolidation has been associated with weaker than projected growth but this. conclusion holds only in some years and only when Greece is included The repeated assumption that the. euro crisis would dissipate over time and that sovereign bond yield differentials would narrow has been a. more important source of error, The forecasting experience in the wake of the crisis has led to a number of changes in forecasting. procedures and communication in the OECD as well as in other forecasting institutions. Forecasting during the crisis, 1 This note discusses OECD forecast performance over the period 2007 12 It focuses on the. lessons that can be learned from cross country differences in growth forecast errors and the changes to. forecasting models and procedures that have been prompted by the experience of the crisis A more. detailed statistical evaluation of OECD forecast performance and information on growth and inflation. projection errors in different economies is provided in the accompanying working paper Pain. et al 2014, The pattern of forecast errors since the crisis began.
2 On average across countries calendar year GDP growth was overestimated across 2007 12. Table 1 Forecasts were revised down consistently and very rapidly when the financial crisis erupted but. growth outturns nonetheless still proved substantially weaker than had been projected The onset of the. euro area crisis in 2010 with the re pricing of sovereign debt and banking sector risks subsequently. contributed to a further period of growth disappointments relative to forecasts. 3 As would be expected errors in projections of current year growth are smaller than in projections. of GDP growth for the following year On average growth was overestimated in both the OECD and. BRIICS economies though the errors in the latter were slightly smaller than in the OECD economies. Figure 1 The largest errors were made in the vulnerable euro area economies. Table 1 Average errors of calendar year GDP growth projections for OECD countries. 2007 12 percentage points, May projection for current November projection for May projection for next. year next year year,Full period 2007 12 0 1 0 9 1 4. Sub periods,2007 09 0 2 1 8 2 6,2010 12 0 1 0 1 0 3. Note Errors are calculated as actual growth less forecast growth at each forecast horizon where actual growth is the published. outturn as at May the following year A negative positive average error indicates over under prediction There is only partial. coverage for Chile Estonia Israel and Slovenia who acceded to the OECD during the period. Source OECD Economic Outlook databases and OECD calculations. Figure 1 Average forecast errors of calendar year GDP growth differ by country group and date of forecast. 2007 12 percentage points, Note Errors are calculated as actual growth less forecast growth at each forecast horizon where actual growth is the published. outturn as at May the following year A negative positive average error indicates over under prediction There is only partial. coverage for Chile Estonia Indonesia Israel Slovenia and South Africa Countries included in the euro area vulnerable group are. Greece Ireland Italy Portugal and Spain, Source OECD Economic Outlook databases and OECD calculations.
Factors correlated with the forecast errors, 4 The growth forecast errors appear to be related to a number of economic features institutional. factors and economic policy developments in the countries concerned Pair wise correlations between the. country forecast errors three broad sets of pre crisis factors and four indicators of contemporaneous. economic developments are set out in Table 2 Key findings include. Growth was comparatively weaker than expected and forecast errors higher in countries that are. more open to external developments and exposed to shocks from other economies Factors such. as openness to international trade and the share of total national banking assets held by. foreign owned banks are negatively correlated with forecast errors over the period as a whole and. particularly so during the downturn Figure 2 This suggests that the projections failed to fully. reflect the increasing globalisation of real and financial activity prior to the crisis which had. raised the potential for cross border and cross market transmission of economic and financial. shocks For instance at the start of the crisis foreign owned banks an increasingly important. pre crisis presence in domestic banking markets often cut credit extensions or reduced new. lending in their host economies in order to meet lower risk targets imposed by their parent banks. Cetorelli and Goldberg 2011 1, Table 2 Growth projection errors reflect a number of economic features. Full period 2007 12 Sub periods,Cumulative error over two years shown. Average error RMSE at publication date indicated, From projections at May for the 2008 09 2010 11 2011 12. next year May 08 EO May 10 EO May 11 EO,Trade openness.
Financial openness,Foreign banks assets,Economy wide regulations. Product market regulation,Employment protection legislation. Financial structure,Stock market capitalisation,Regulatory capital. Financial services,Contemporaneous developments,Equity price growth. Change in non performing loans,Change in business confidence manufacturing.
Change in consumer confidence, Note RMSE denotes the root mean squared error of the growth projections denotes variables with extreme values that have been. omitted because they change the nature of the relationship see Pain et al 2014 for details The nature of each pairwise. correlation is shown by a positive or negative sign the strength is shown by one two or three signs indicating statistical significance. at the 10 5 or 1 level respectively with statistically insignificant relationships omitted See Table 3 in Pain et al 2014 for full. results Contemporaneous developments are the change over the relevant period being considered. Source Datastream IMF Financial Soundness Indicators OECD Economic Outlook databases OECD Going for Growth 2012. OECD Main Economic Indicators database OECD Structural Analysis database World Bank Global Financial Development. database and OECD calculations, 1 The evidence points to the enhanced exposure of host economies to source economy shocks during the downturn. outweighing the potential for foreign owned banks with access to internal capital resources to help shield host economies. from domestic shocks which would limit negative growth surprises. Figure 2 Growth projection errors for 2008 09 were larger in countries with more international openness. Cumulative errors for 2008 09 calendar year projections made at May 2008. Note Errors are calculated as actual growth less forecast growth at each forecast horizon where actual growth is the published. outturn as at May the following year A negative positive error indicates over under prediction Countries with partial coverage for. the period Chile Estonia Israel and Slovenia are excluded. Source OECD Economic Outlook databases World Global Financial Development database and OECD calculations. Figure 3 Forecast errors were larger in more regulated economies. Average RMSE of annual growth projections for 2007 to 2012 made in November a year earlier. Note RMSE denotes the root mean squared error of the growth projections Countries with partial coverage for the period Chile. Estonia Israel and Slovenia are excluded, Source OECD Going for Growth 2012 OECD Economic Outlook databases and OECD calculations. Larger forecast errors over 2007 12 have occurred in countries with more stringent pre crisis. labour and product market regulations Figure 3 In part this may reflect the weight given at the. time to pre crisis evidence that tight regulations could help to cushion economic shocks Duval. et al 2007 together with insufficient attention being paid to the extent to which tighter. regulations could delay necessary reallocations across sectors in the recovery phase A third. possibility is that it reflects a correlation between restrictive regulations and the pre crisis. build up of imbalances that was not fully captured in forecasts 2. Growth has been weaker than expected in the recovery in countries where banks had a low ratio. of regulatory capital to risk weighted assets in 2007 Figure 4 left panel This may suggest that. insufficient account was taken of the impact of the rapid pre crisis expansion of financial. leverage in a low interest rate environment and the greater need for weakly capitalised banks to. deleverage subsequently, There are strong correlations between forecast errors and changes in equity prices private sector. confidence and non performing loans in the banking sector Figure 4 right panel during the. post crisis recovery period This suggests that the impacts of impaired banking systems and weak. confidence may have been underestimated in the projections 3 There is evidence that these. variables were also correlated with forecast errors in the pre crisis period Pain et al 2014. 5 Perhaps surprisingly indicators of pre crisis vulnerabilities such as house price increases and. private credit growth appear not to be generally associated with negative growth surprises since the crisis. began A possible explanation might be that these vulnerabilities were generally well understood and. incorporated successfully into forecast judgements Not all pre crisis imbalances were equally well. integrated in the projections however with higher downside surprises on average in countries that had a. pre crisis current account deficit, Fiscal consolidation bond spreads and growth disappointments in 2010 11 and 2011 12.
6 In 2010 12 a repeatedly projected acceleration in the pace of the global recovery ultimately. failed to materialise As mentioned above an underestimate of the impact of impaired banking systems and. bank deleveraging may be one factor behind this Figure 4 These years also saw the deepening of the. euro area crisis and the onset of widespread fiscal consolidation Indeed IMF studies point to a negative. relationship between the errors in its Spring 2010 World Economic Outlook projections for GDP growth. over 2010 11 and projected fiscal consolidation during this period IMF 2012 Blanchard and Leigh. 2013 An implication drawn from this is that fiscal multipliers were underestimated considerably. 7 A more nuanced picture appears from the OECD projections published in the Economic Outlook. in May 2010 and May 2011 for cumulative GDP growth during 2010 11 and 2011 12 respectively. Projected fiscal consolidation is negatively correlated with the growth forecast errors in 2010 11 Figure 5. upper left panel but not in 2011 12 Figure 5 lower left panel Under the hypothesis that the former. reflects an initial underestimation of the fiscal multiplier the latter would then suggest that OECD. forecasters became better at judging the impact of consolidation in an environment with limited monetary. policy space However the finding for 2010 11 is not statistically significant without the inclusion of. Greece And among non European countries there is no clear association at all between projected. consolidation and forecast errors raising doubts about the hypothesis of a generalised underestimation of. fiscal multipliers, 2 Related measures of business and labour market regulation produced by the World Bank and the Fraser Institute are found. OECD FORECASTS DURING AND AFTER THE FINANCIAL CRISIS A POST MORTEM Main Findings GDP growth was overestimated on average across 200712 reflecting not only errors at the height of the financial crisis but also errors in the subsequent recovery

Related Books