Slovakia s early experience with the euro during global recession. Jan Toth. Deputy Governor National Bank of Slovakia

Transcription

1 Slovakia s early experience with the euro during global recession Jan Toth Deputy Governor National Bank of Slovakia

2 Introductory caveat Ahead of its euro changeover, Slovakia lacked nominal exchange rate flexibility from H This coincided, however, with the outbreak of the Great Recession The Great Recession was a unique event whose impact on Slovakia far outweighed that of the euro adoption Given the unusual scale of this external shock, Slovakia s experiences over the past seven years may yet be far different from long-run trends 2

3 Summary: Slovakia s early experience with the euro Materialisation of immediate benefits was mostly in line with assumptions: transaction and other costs declined (by 0.3% of GDP) Assumed longer-term benefits of 7-20% of GDP still to materialise Euro area debt crisis induced additional costs related to contagion and contributions to the EFSF and ESM Real exchange rate not stable - quite the opposite Larger adjustment through employment Currently weak real exchange rate provides additional cyclical stimulus Still no clear sign of significantly larger longer-term benefits (higher FDI, trade growth, potential GDP growth) 3

4 Euro adoption coinciding with the global crisis Great Recession impact adjustment is aided by making a comparison with a similar crisis-hit economy (Czech Republic) Global shock induced by the economic crisis highlighted the lack of (nominal) exchange rate flexibility Real exchange rate stability not achieved, as important trading partners saw their local currencies fluctuate to buffer the external shock In fact, the lack of a national currency increased real exchange rate volatility 4

5 Outline Output and employment Inflation Foreign trade and investment Economic flexibility and structural reforms Euro debt crisis 5

6 Output and employment 6

7 Is it possible to explain SK growth outperformance by differences in fiscal policies? Average GDP growth (%, p.p.) - decomposition Average GDP growth (%,p.p.) - decomposition 7,0 5,7 6,0 4,5 5,0 Fiscal policy differences in do not suffice to explain growth differential between SK and CZ Fiscal stance defined as change in primary structural balance multiplied by multiplier (0.6). Sources: IMF WEO 10/2015. ESA2010 data 4,0 3,0 2,0 1,0 0,0-1,0-2,0 2,0 0,1 0,6-1,1 CZ SK CZ SK CZ SK F impulse GDP excl. F impulse GDP It may, however, be better to focus specifically on exports/manufacturing sectors and relative changes in GDP forecasts (as opposed to absolute levels). 7

8 Due to exports, both the recession and recovery were more marked in Slovakia than in the Czech Republic Comparison of exports and foreign demand dynamics [in %] SK CZ SK CZ SK CZ SK CZ SK CZ SK CZ SK CZ Exports Foreign demand Market share Source: ECB, Eurostat, NBS calculation. 8

9 Employment adjustment in manufacturing stronger in EA Countries Employment - manufacturing (Index: 2008Q2=100) Q2 2009Q2 2010Q2 2011Q2 2012Q2 2013Q2 2014Q2 2015Q2 CZ HU PL SI SK Note: CZ, PL, SK seasonally adjusted data; HU, SI seasonally and working day-adjusted data. SI and SK are members of the euro area. Sources: Eurostat, NBS calculations. 9

10 % change in compensation per employee % change in compensation per employee In SK and SI, where wage flexibility was lower due to euro adoption, firms responded to crises by cutting jobs Wage flexibility similar in local currency Wage flexibility in EUR lower in SK and SI Change in employment and compensation per employee in local currency in manufacturing (2008Q2-2009Q4) Change in employment and compensation per employee in EUR in manufacturing (2008Q2-2009Q4) 12,0 12,0 6,0 SI PL 6,0 SI 0,0 SK CZ HU 0,0 SK -6,0-6,0-12,0-12,0 CZ HU PL -18,0-16,0-14,0-12,0-10,0-8,0-6,0-4,0 % change in employment -18,0-16,0-14,0-12,0-10,0-8,0-6,0-4,0 % change in employment Sources: Eurostat, NBS calculations. Sources: Eurostat, NBS calculations. 10

11 Surprises in the early stage of the crisis Real output undershot forecasts by a cumulative 10.7% in the Czech Republic and an even higher 15.3% in Slovakia during GDP - cumulative forecast error [in p.p.] CZ growth forecast Actual GDP growth Difference Source: CNB forecast (7/8/2008) SK growth forecast Actual GDP growth Difference Source: NBS forecast (P3Q-2008) CZ SK Sources: NBS, CNB, NBS calculation 11

12 Surprises in the later stages of the crisis But the later negative surprises in Slovakia were smaller Recent recovery in Slovakia has been stronger than projected (-4.1% in CZ and -2.3% in SK) GDP cumulative forecast error [in pp] CZ growth forecast 1,8 2,9 Actual GDP growth 1,8-1,2 Difference 0-4,1 Source: CNB forecast (5/8/2010) SK growth forecast 3,0 4,1 Actual GDP growth 3,2 2,0 Difference +0,2-2,1 Source: NBS forecast (P3Q-2010) 0,5 0,0-0,5-1,0-1,5-2,0-2,5-3,0-3,5-4,0-4, CZ SK Sources: NBS, CNB, NBS calculations. 12

13 Q1 1998Q1 1999Q1 2000Q1 2001Q1 2002Q1 2003Q1 2004Q1 2005Q1 2006Q1 2007Q1 2008Q1 2009Q1 2010Q1 2011Q1 2012Q1 2013Q1 2014Q1 2015Q1 Exchange rate today is probably undervalued Historically high cyclically-adjusted current account/trade surpluses 10 Net exports of goods and services [in % of real GDP, 5-quarter moving average] 4 Cyclically adjusted Current Account [in % of nominal GDP, 5-quarter moving average] Note: BPM5 methodology since 2008 BPM6 methodology Cyclically adjusted net exports estimates band Net exports Cyclically adjusted Current Account to GDP ratio Note: Cyclically-adjusted net exports were constructed from a comparison of FEER model simulations using actual and cyclicallyadjusted paths of labour productivity differentials, foreign demand and domestic demand. The estimates band is a result of three alternative cyclically-adjusted domestic demand estimates: an equilibrium path from the FEER approach, a trend component approximated through the output gap, and domestic demand smoothed with a Hodrick-Prescott filter. The cyclical component of the current account to GDP ratio was calculated using its elasticity on the foreign and domestic output gap differential estimate of 0.28 in Haltmaier J. (2014) Cyclically Adjusted Current Account Balances, Board of Governors of the Federal System, International Finance Discussion Papers Number Source: ECB, OECD, NBS calculation. 13

14 Slovak real exchange rate has been more volatile (higher overvaluation and greater undervaluation) 1997Q3 1998Q3 1999Q3 2000Q3 2001Q3 2002Q3 2003Q3 2004Q3 2005Q3 2006Q3 2007Q3 2008Q3 2009Q3 2010Q3 2011Q3 2012Q3 2013Q3 2014Q3 10 manufacturing PPI-based REER misalignment [in %] Overvaluation Undervaluation -4-6 SK CZ Notes: The chart shows five-quarter centered moving averages; CZK intervention floor from 2013Q4; CZ bilateral RER CZK/EUR Sources: NBS, CNB. 14

15 1997Q1 1998Q1 1999Q1 2000Q1 2001Q1 2002Q1 2003Q1 2004Q1 2005Q1 2006Q1 2007Q1 2008Q1 2009Q1 2010Q1 2011Q1 2012Q1 2013Q1 2014Q1 2015Q1 Real exchange rate undervaluation Historically, in the local currency, wage growth in SK was twice as strong as in DE, while in euro it was three times as strong. After the euro adoption and the start of the Great Recession, labour costs (nominal compensation per employee) have been rising at a similar pace. The impossibility of nominal appreciation (together with low inflation) leads to an undervalued real exchange rate that helps Slovakia to gain world trade market share. Compensation of employees 20,0 15,0 10,0 5, Predikcia v číslach Real exchange rate overvaluation/undervaluation Equilibrium band REER equilibrium Overvaluation 0,0-5,0-10, Slovakia (EUR, fixed conversion rate) Slovakia (EUR) Germany (EUR) Sources: Eurostat, NBS calculations. Note: annual percentage changes; Q4 monthly average; seasonallyadjusted data in current prices. Source: NBS. Note: manufacturing PPI-based REER, equilibrium, corresponding to equilibrium exchange rate projection in the MTF-2015Q3 medium-term forecast + appreciation, - depreciation. 15

16 Early in the crisis, the exchange rate procyclically weighed on economic growth; later it countercyclically boosted it 2004Q1 2005Q1 2006Q1 2007Q1 2008Q1 2009Q1 2010Q1 2011Q1 2012Q1 2013Q1 2014Q REER misalignment and market share [in p.p. and %] Production capacity increase related to the FDI inflow (nonprice effect) Overvaluation -6-8 Annual changes of REER misalignment Note: manufacturing PPI-based REER; five-quarter centered moving averages. Annual changes of market share Sources: ECB, Eurostat, NBS calculations. 16

17 1998Q4 1999Q4 2000Q4 2001Q4 2002Q4 2003Q4 2004Q4 2005Q4 2006Q4 2007Q4 2008Q4 2009Q4 2010Q4 2011Q4 2012Q4 2013Q4 2014Q4 Historically huge procyclicality REER misalignment and output gap (right-hand scale) [in %] Procyclical REER impact Overvaluation REER misalignment Output gap (right-hand scale) Note: PPI-manufacturing based REER; five-quarter centered moving averages. Source: NBS calculations. 17

18 What if - illustrative model scenarios Counterfactuals: Short-run approach: currency copies nominal movements of CZK Medium-run approach: currency copies real movements of CZK Long-run approach: currency is always in equilibrium 18

19 SKK nominal changes equal to CZK changes (short-run approach) Difference from the outcome level [in %] Sources: Eurostat, NBS calculations. 19

20 SKK and CZK trends equal in real terms (mediumrun approach) Difference from the outcome level [in %] Sources: Eurostat, NBS calculations. 20

21 SKK always in (real) equilibrium (long-run approach) Difference from the outcome level [in %] Sources: Eurostat, NBS calculations. 21

22 Illustrative model scenarios w/o euro Short-run view recession milder by 2 p.p., positive difference remains Medium-run view recession milder by 1 p.p., positive difference disappears Longer-run view recession milder by 1 p.p., positive difference disappears 22

23 Inflation 23

24 GDP per capita in PPP (2012, EA=100) No grounds yet to support concerns about inflation Despite expected price convergence SK inflation close to EA average 130 Inflation vs GDP per capita 120 IE DE AT EA FI MT GR LV SK LT EE 50 0,0 0,5 1,0 1,5 2,0 2,5 average annual HICP inflation ( M9) Source: Eurostat 24

25 GDP per capita in PPP (2012, EA=100) The same goes for core inflation (so don t blame energy and food) Core inflation vs GDP per capita IE DE NL BE FI AT FR EA CY ES IT GR SI PT LT MT SK EE 60 LV Source: Eurostat 50-0,5 0,0 0,5 1,0 1,5 2,0 average annual core inflation (Overall HICP excluding energy, food, alcohol and tobacco; M9) 25

26 (Core) inflation during euro period: so far very low Initially, the lagged impact of exchange rate appreciation prior to euro adoption pushed inflation down in In addition, service prices were kept low by weak domestic demand (global recession). Changeover impact estimated to no more than 0.2% - a non-issue. HICP inflation pre-crisis forecast and actual development Services inflation and tradable goods inflation forecast and actual development Exchange rate appreciation, global crisis and collapse of domestic demand HICP forecast HICP Non-energy industrial goods forecast Services forecast Non-energy industrial goods Services Sources: NBS, Eurostat. Sources: NBS, Eurostat. 26

27 Low CZ inflation became even lower after 2009 Tradables inflation and services inflation very similar in SK and CZ during CZ inflation has since been lower. Non-energy industrial goods at constant tax rates (index: 2005 = 100), SK vs CZ Services at constant tax rates (index: 2005 =100, SK vs CZ Source: Eurostat. Source: Eurostat. 27

28 Higher inflation in SK created larger buffer for real interest rates to adjust in ZLB environment. 2000Q1 2001Q1 2002Q1 2003Q1 2004Q1 2005Q1 2006Q1 2007Q1 2008Q1 2009Q1 2010Q1 2011Q1 2012Q1 2013Q1 2014Q1 2015Q1 4 3M real interest rates (in %) SK CZ Note: 3M MM deflated by HICP excluding food and energy prices; three-quarter centered moving averages. Source: ECB, OECD, NBS calculations. 28

29 Foreign trade and investment 29

30 Market shares (annual percentage changes) - Situation in SK was worse at the start of the crisis, but better later on Sources: Eurostat, ECB, NBS calculations. 30

31 Trade trends still in line with assumptions for long-run gains from euro (but CZ also performs well) Export growth observed also in Czech Republic, so not necessarily euro-specific. Bigger improvement recorded in non-ea exports. Actual vs expected intra-ea exports (SK / CZ) with interval bands assumed in the NBS calculation of longer-term benefits of euro adoption (extra export growth due to common currency) 70 40% Contributions to real export growth since Jan % % 50 Czech Republic 10% % -10% Extra EU Slovak Republic -20% -30% Non-EA EA Total Sources: Eurostat, NBS calculations. Sources: Eurostat, NBS calculations. 31

32 Impact of global trade developments on exports SK export performance improved. Despite euro adoption, exports to extra-ea countries picked up (in terms of both volume and market shares) Average YoY growth [%] SK Real exports Total exports Average YoY growth [%] CZ Real exports Total export Intra EA Extra EA Average YoY growth [%] Market shares Total Intra Average YoY growth [%] Market shares Total Extra Sources: Eurostat, NBS calculations. 32

33 Foreign direct investment (% of GDP) For FDI inflows, Slovakia s results were slightly better before euro adoption and more balanced/mixed afterwards. FDI inflows / GDP (3Y MA) SK CZ 33

34 Economic flexibility and structural reforms 34

35 Economic flexibility / Structural reforms Non-EMU countries growth stronger than EA average in recent years To mitigate nominal exchange rate inflexibility, the economy should be reformed so that other parts are as flexible as possible Has EA carried out further structural reforms to offset the lack of nominal currency flexibility? 35

36 Change in Change in Is reform process pace slightly stronger in EA? The EA s ease of doing business ranking is slightly higher due in part to reforms forced upon programme countries. In this respect, both SK and CZ are lagging. Entire period since crisis Since GR PT SI PL HR GR PT SI PL ES IT HR CZ SK -20 SK CZ Ranking in Ranking in

37 New Zealand Saudi Arabia United States Canada United Kingdom Brazil Chile Australia South Africa Estonia Ireland Hungary Japan Switzerland Finland Korea Israel Slovak Republic Spain OECD average Norway Denmark Poland Greece Austria Iceland Turkey Russian Federation Sweden India Mexico Czech Republic Slovenia Portugal Luxembourg Italy France Argentina Latvia Netherlands Indonesia Belgium Germany China Labour market flexibility in Slovakia Countries without their own currency should have above-average labour market flexibility (LBF). LBF in Slovakia improved markedly, but remains only close to OECD average. 3,5 EPL Index (eprc_v3) OECD countries 3,0 EPL Index - Slovakia More rigid 3,0 2,8 2,5 2,0 2,6 1,5 1,0 2,4 0,5 2,2 0,0 2, SK CZ 37

38 Euro debt crisis 38

39 10Y-Interest Rate Sovereign debt vs debt costs OECD+EU countries - weak relationship between debt costs and indebtedness Debt/GDP Sources: OECD. Eurostat Note: Averages for Estimates for 2015 by OECD. 39

40 10Y-Interest Rate 10Y-Interest Rate Sovereign debt vs debt costs The relationship seems to be more pronounced for the euro area (no local central bank as a lender of last resort). OMT helped, but differences remain. Before OMT After OMT Euro countries 10 Euro countries Non-euro countries 4 Non-euro countries Debt/GDP Debt/GDP Sources: OECD, EC, Eurostat. Note: The average for non-euro area countries shown in both charts. In the case of euro area countries, the pre-omt period represents the average for and the post-omt period corresponds to

41 Non investment grade +1 Investment grade Rating downgrades for Slovakia and other EA countries EMU membership specifically mentioned in explanation of downgrade in 2012 (EA countries downgraded en bloc) Rating S&P Grade AA- CZ CZ A+ SK SK A CZ SK A- CZ SK = CZ BBB+ CZ BBB BBB- BB+ BB BB- SK -5 grades SK -4 grades

42 SK CZ bond yield spread During most of the crisis, spreads widened sharply, but recently they have moved back towards pre-euro levels. 10Y bond yields in SK bond yields in CZ (in p. p.) 2,50 2,00 Euro adoption EA debt crisis 1,50 1,00 0,50 0,00-0,50 Source: ECB. 42

43 Some lessons learned Materialisation of immediate benefits of euro was mostly in line with assumptions: transaction costs declined, changeover effect low EA debt crisis induced additional costs related to contagion and fiscal contributions to the EFSF and ESM Huge external shock highlighted lack of exchange rate flexibility vis-a-vis neighbours Lack of a currency buffer caused currency to became strongly overvalued during recession Later, a weaker real exchange rate seems to have provided necessary additional stimulus (SK not in recession in 2013) Still no clear sign of the most important longer-term benefits (higher FDI and trade growth). It is vital that the rest of the economy be as flexible as possible in order to offset exchange rate rigidity 43

44 Appendix 44

45 Slovak price level converged, unlike Czech price level All CE-4 countries apart from Slovakia saw price-level divergence during the crisis. Convergence in Slovakia explained mainly by prices of hotels/restaurants and food. Price level comparison with EA and price-level change relative to EA countries in 2008 and 2014 Price-level change relative to EA countries in 2008 and 2014 Price-level change relative to EA countries in 2008 and 2014 Actual individual consumption Hotels and restaurants Food and non-alcoholic beverages 45

46 What do prices of frequent out-of-pocket purchases imply? Not much. The inflation rate for frequent out-of-pocket purchases (FROOPP) did not change significantly after the euro adoption, despite popular perception to the contrary. Even the infamous price of coffee has not rocketed (though it did increase somewhat just before the euro adoption). Comparison of annual rate of change in FROOPP inflation, excluding administered prices and cup of coffee price. FROOPP inflation in SK and EA (annual percentage change) Cup of coffee (price level and annual percentage change) 46