Discussion of Income-Induced Expenditure Switching

Transcription

1 Discussion of Income-Induced Expenditure Switching Rudolfs Bems & Julian di Giovanni Franck Portier Toulouse School of Economics CEPR Conference on heterogeneity in currency areas and macroeconomic policies ECB / 19

2 0. Roadmap 1. Macroeconomic context 2. The puzzle 3. Income driven expenditure switching 4. Epirical analysis 5. In a nutshell 6. Where to go? 2 / 19

3 0. Roadmap 1. Macroeconomic context 2. The puzzle 3. Income driven expenditure switching 4. Epirical analysis 5. In a nutshell 6. Where to go? 2 / 19

4 0. Roadmap 1. Macroeconomic context 2. The puzzle 3. Income driven expenditure switching 4. Epirical analysis 5. In a nutshell 6. Where to go? 2 / 19

5 0. Roadmap 1. Macroeconomic context 2. The puzzle 3. Income driven expenditure switching 4. Epirical analysis 5. In a nutshell 6. Where to go? 2 / 19

6 0. Roadmap 1. Macroeconomic context 2. The puzzle 3. Income driven expenditure switching 4. Epirical analysis 5. In a nutshell 6. Where to go? 2 / 19

7 0. Roadmap 1. Macroeconomic context 2. The puzzle 3. Income driven expenditure switching 4. Epirical analysis 5. In a nutshell 6. Where to go? 2 / 19

8 1. Macroeconomic context Latvia went through a brutal boom-bust-recovery episode since (Blanchard, Griffiths and Gruss) An increase in GDP of almost 90% from 2000Q1 to 2007Q4, a decrease of 25% from 2007Q4 2 to 2009Q3, A recovery, as of 2013Q1, of 18%. 2007:4, followed by an increase to more than 21% in 2010:1, and a decrease since then, down to 11.4% in 2013:2. Figure 1: GDP and Unemployment in Latvia Figure 1. Latvia: Real GDP and the Unemployment Rate (sa), Unemployment (lhs) Real GDP (2000q1=100, rhs) 3 / 19

9 1. Macroeconomic context Latvia went through a brutal boom-bust-recovery episode since (Blanchard, Griffiths and Gruss) An increase in GDP of almost 90% from 2000Q1 to 2007Q4, a decrease of 25% from 2007Q4 2 to 2009Q3, A recovery, as of 2013Q1, of 18%. 2007:4, followed by an increase to more than 21% in 2010:1, and a decrease since then, down to 11.4% in 2013:2. Figure 1: GDP and Unemployment in Latvia Figure 1. Latvia: Real GDP and the Unemployment Rate (sa), Unemployment (lhs) Real GDP (2000q1=100, rhs) 3 / 19

10 1. Macroeconomic context Latvia went through a brutal boom-bust-recovery episode since (Blanchard, Griffiths and Gruss) An increase in GDP of almost 90% from 2000Q1 to 2007Q4, a decrease of 25% from 2007Q4 2 to 2009Q3, A recovery, as of 2013Q1, of 18%. 2007:4, followed by an increase to more than 21% in 2010:1, and a decrease since then, down to 11.4% in 2013:2. Figure 1: GDP and Unemployment in Latvia Figure 1. Latvia: Real GDP and the Unemployment Rate (sa), Unemployment (lhs) Real GDP (2000q1=100, rhs) 3 / 19

11 1. Macroeconomic context Latvia went through a brutal boom-bust-recovery episode since (Blanchard, Griffiths and Gruss) An increase in GDP of almost 90% from 2000Q1 to 2007Q4, a decrease of 25% from 2007Q4 2 to 2009Q3, A recovery, as of 2013Q1, of 18%. 2007:4, followed by an increase to more than 21% in 2010:1, and a decrease since then, down to 11.4% in 2013:2. Figure 1: GDP and Unemployment in Latvia Figure 1. Latvia: Real GDP and the Unemployment Rate (sa), Unemployment (lhs) Real GDP (2000q1=100, rhs) 3 / 19

12 1. Macroeconomic context Since 2005, the Bank of Latvia is following a strict policy of pegging the lats to a basket composed of the US dollar, the euro, the pound sterling and the Japanese yen. Gnuplot Figure 2: Euro-Lats exchange rate 27/11/ : Euro in LVL LVL / 19

13 1. Macroeconomic context The recovery has been made possible by internal devaluation fiscal austerity 5 / 19

14 of 2009, ULCs had declined by close to 25%, and they have remained roughly stable since. While wage cuts played a role initially, over time the reduction in ULCs has come entirely from productivity improvements. 1. Macroeconomic context Figure 9. Cumulative Change in Wages, Productivity and ULCs Whole Economy since 2008:3 (2008q3=100) Figure 3: Internal devaluation - Total economy (cumulative change) ULC Output per occupied post Wage per occupied post 70 6 / 19

15 manufacturing, have continued. Looking across subsectors within manufacturing, productivity continued increasing even in subsectors where employment growth has resumed. 1. Macroeconomic context Figure 10. Cumulative Change in Wages, Productivity and ULCs Manufacturing since 2008:3 (2008q3=100) Figure 4: Internal devaluation - Manufacturing (cumulative change) ULC Output per occupied post Wage per occupied post / 19

16 The first is the increase in the headline deficit in 2009; but given the very large decline in output, this corresponded to a reduction in the cyclically adjusted deficit. 1. Macroeconomic context Figure 8. Fiscal Impulse: Change in Headline and Cyclically Adjusted Balances Figure 5: Fiscal austerity (percent of GDP, using latest EC output gap estimate) Headline (change, excl. bank restructuring costs) Cyclically Adjusted (change, excl. bank restructuring costs) Cyclically Adjusted (change, incl. bank restructuring costs) 8 / 19

17 by a large shift towards foreign goods, leading to an even larger deterioration of the current account balance. (Figure 2 plots the evolution of exports, imports, and the trade balance). Part of the shift probably reflected the increasing real exchange rate appreciation (more on this below). Part of it may have also reflected a shift towards higher quality foreign products, an issue relevant when we look at the current account adjustment later Macroeconomic context After years of large trade deficit, the recession has been accompanied by a sharp trade balance recovery Figure 2. Latvia: Trade Balance, Figure 6: Trade balance (national accounts basis, 4 quarter moving sum, percent of GDP) Trade Balance Exports (rhs) Imports (rhs) 9 / 19

18 2. The puzzle In open macroeconomics, an improvement of the trade balance in recessions is typically a consequence of a decrease in imports caused but the decrease of absorption during the recession an expenditure switching (demand shifts from foreign goods to domestic ones) that is caused by a depreciation of the exchange rate. The expenditure switching channel should not be seen in the Latvian context as There was no external devaluation The internal devaluation was only partly transmitted to prices, leading more to an increase in margins for firms. (quoting Blanchard et al.) Why was the current account improvement then so large? 10 / 19

19 2. The puzzle In open macroeconomics, an improvement of the trade balance in recessions is typically a consequence of a decrease in imports caused but the decrease of absorption during the recession an expenditure switching (demand shifts from foreign goods to domestic ones) that is caused by a depreciation of the exchange rate. The expenditure switching channel should not be seen in the Latvian context as There was no external devaluation The internal devaluation was only partly transmitted to prices, leading more to an increase in margins for firms. (quoting Blanchard et al.) Why was the current account improvement then so large? 10 / 19

20 2. The puzzle In open macroeconomics, an improvement of the trade balance in recessions is typically a consequence of a decrease in imports caused but the decrease of absorption during the recession an expenditure switching (demand shifts from foreign goods to domestic ones) that is caused by a depreciation of the exchange rate. The expenditure switching channel should not be seen in the Latvian context as There was no external devaluation The internal devaluation was only partly transmitted to prices, leading more to an increase in margins for firms. (quoting Blanchard et al.) Why was the current account improvement then so large? 10 / 19

21 2. The puzzle In open macroeconomics, an improvement of the trade balance in recessions is typically a consequence of a decrease in imports caused but the decrease of absorption during the recession an expenditure switching (demand shifts from foreign goods to domestic ones) that is caused by a depreciation of the exchange rate. The expenditure switching channel should not be seen in the Latvian context as There was no external devaluation The internal devaluation was only partly transmitted to prices, leading more to an increase in margins for firms. (quoting Blanchard et al.) Why was the current account improvement then so large? 10 / 19

22 2. The puzzle In open macroeconomics, an improvement of the trade balance in recessions is typically a consequence of a decrease in imports caused but the decrease of absorption during the recession an expenditure switching (demand shifts from foreign goods to domestic ones) that is caused by a depreciation of the exchange rate. The expenditure switching channel should not be seen in the Latvian context as There was no external devaluation The internal devaluation was only partly transmitted to prices, leading more to an increase in margins for firms. (quoting Blanchard et al.) Why was the current account improvement then so large? 10 / 19

23 2. The puzzle In open macroeconomics, an improvement of the trade balance in recessions is typically a consequence of a decrease in imports caused but the decrease of absorption during the recession an expenditure switching (demand shifts from foreign goods to domestic ones) that is caused by a depreciation of the exchange rate. The expenditure switching channel should not be seen in the Latvian context as There was no external devaluation The internal devaluation was only partly transmitted to prices, leading more to an increase in margins for firms. (quoting Blanchard et al.) Why was the current account improvement then so large? 10 / 19

24 2. The puzzle In open macroeconomics, an improvement of the trade balance in recessions is typically a consequence of a decrease in imports caused but the decrease of absorption during the recession an expenditure switching (demand shifts from foreign goods to domestic ones) that is caused by a depreciation of the exchange rate. The expenditure switching channel should not be seen in the Latvian context as There was no external devaluation The internal devaluation was only partly transmitted to prices, leading more to an increase in margins for firms. (quoting Blanchard et al.) Why was the current account improvement then so large? 10 / 19

25 3. Income driven expenditure switching Assume that consumption of domestic and foreign products are given by C d = ( ) Y εd y ep ε d e P C f = ( ) Y εf y ep ε f e P with all elasticities being positive. Hometheticity of preferences implies ε d y = ε f y so that C d C f = ( ep P ) ε d e +ε d e With homothetic preferences, there is no expenditure switching absent of variations in the real exchange rate. 11 / 19

26 3. Income driven expenditure switching Assume that consumption of domestic and foreign products are given by C d = ( ) Y εd y ep ε d e P C f = ( ) Y εf y ep ε f e P with all elasticities being positive. Hometheticity of preferences implies ε d y = ε f y so that C d C f = ( ep P ) ε d e +ε d e With homothetic preferences, there is no expenditure switching absent of variations in the real exchange rate. 11 / 19

27 3. Income driven expenditure switching Assume that consumption of domestic and foreign products are given by C d = ( ) Y εd y ep ε d e P C f = ( ) Y εf y ep ε f e P with all elasticities being positive. Hometheticity of preferences implies ε d y = ε f y so that C d C f = ( ep P ) ε d e +ε d e With homothetic preferences, there is no expenditure switching absent of variations in the real exchange rate. 11 / 19

28 3. Income driven expenditure switching Assume that consumption of domestic and foreign products are given by C d = ( ) Y εd y ep ε d e P C f = ( ) Y εf y ep ε f e P with all elasticities being positive. Hometheticity of preferences implies ε d y = ε f y so that C d C f = ( ep P ) ε d e +ε d e With homothetic preferences, there is no expenditure switching absent of variations in the real exchange rate. 11 / 19

29 3. Income driven expenditure switching Rudolfs and Julian find a substantial amount of expenditure switching. This is possible with non nomothetic preferences Assume that domestic goods are inferior compared to the foreign ones ε d y < ε f y then C d C f ( = Y εd y εf y ep P ) ε d e +ε d e With non nomothetic preferences i.e. ε d y < ε f y, there is expenditure switching when income goes down. 12 / 19

30 3. Income driven expenditure switching Rudolfs and Julian find a substantial amount of expenditure switching. This is possible with non nomothetic preferences Assume that domestic goods are inferior compared to the foreign ones ε d y < ε f y then C d C f ( = Y εd y εf y ep P ) ε d e +ε d e With non nomothetic preferences i.e. ε d y < ε f y, there is expenditure switching when income goes down. 12 / 19

31 3. Income driven expenditure switching Rudolfs and Julian find a substantial amount of expenditure switching. This is possible with non nomothetic preferences Assume that domestic goods are inferior compared to the foreign ones ε d y < ε f y then C d C f ( = Y εd y εf y ep P ) ε d e +ε d e With non nomothetic preferences i.e. ε d y < ε f y, there is expenditure switching when income goes down. 12 / 19

32 3. Income driven expenditure switching Rudolfs and Julian find a substantial amount of expenditure switching. This is possible with non nomothetic preferences Assume that domestic goods are inferior compared to the foreign ones ε d y < ε f y then C d C f ( = Y εd y εf y ep P ) ε d e +ε d e With non nomothetic preferences i.e. ε d y < ε f y, there is expenditure switching when income goes down. 12 / 19

33 3. Income driven expenditure switching Rudolfs and Julian find a substantial amount of expenditure switching. This is possible with non nomothetic preferences Assume that domestic goods are inferior compared to the foreign ones ε d y < ε f y then C d C f ( = Y εd y εf y ep P ) ε d e +ε d e With non nomothetic preferences i.e. ε d y < ε f y, there is expenditure switching when income goes down. 12 / 19

34 4. Empirical analysis Rudolfs and Julian do an amazing analysis of scanner-level data for food and beverages and find : imports contraction is by 1/3 accounted for by expenditure switching accounting, while the relative price of foreign goods increased by 4.4% only. switching took place within narrowly defined product groups, while the relative price adjustment was across product groups. Within a category, unit values of domestic goods were on average lower than those of comparable foreign ones. They estimate a demand system with possible non homotheticity and found that there is indeed an relative inferiority of domestic goods. 13 / 19

35 4. Empirical analysis Rudolfs and Julian do an amazing analysis of scanner-level data for food and beverages and find : imports contraction is by 1/3 accounted for by expenditure switching accounting, while the relative price of foreign goods increased by 4.4% only. switching took place within narrowly defined product groups, while the relative price adjustment was across product groups. Within a category, unit values of domestic goods were on average lower than those of comparable foreign ones. They estimate a demand system with possible non homotheticity and found that there is indeed an relative inferiority of domestic goods. 13 / 19

36 4. Empirical analysis Rudolfs and Julian do an amazing analysis of scanner-level data for food and beverages and find : imports contraction is by 1/3 accounted for by expenditure switching accounting, while the relative price of foreign goods increased by 4.4% only. switching took place within narrowly defined product groups, while the relative price adjustment was across product groups. Within a category, unit values of domestic goods were on average lower than those of comparable foreign ones. They estimate a demand system with possible non homotheticity and found that there is indeed an relative inferiority of domestic goods. 13 / 19

37 4. Empirical analysis Rudolfs and Julian do an amazing analysis of scanner-level data for food and beverages and find : imports contraction is by 1/3 accounted for by expenditure switching accounting, while the relative price of foreign goods increased by 4.4% only. switching took place within narrowly defined product groups, while the relative price adjustment was across product groups. Within a category, unit values of domestic goods were on average lower than those of comparable foreign ones. They estimate a demand system with possible non homotheticity and found that there is indeed an relative inferiority of domestic goods. 13 / 19

38 4. Empirical analysis Rudolfs and Julian do an amazing analysis of scanner-level data for food and beverages and find : imports contraction is by 1/3 accounted for by expenditure switching accounting, while the relative price of foreign goods increased by 4.4% only. switching took place within narrowly defined product groups, while the relative price adjustment was across product groups. Within a category, unit values of domestic goods were on average lower than those of comparable foreign ones. They estimate a demand system with possible non homotheticity and found that there is indeed an relative inferiority of domestic goods. 13 / 19

39 5. In a nutshell Consider the (fictious) product group alcoholic beverage with bubbles Pre-crisis basket : 14 / 19

40 5. In a nutshell I Consider the (fictious) product group alcoholic beverage with bubbles I Pre-crisis basket : 14 / 19

41 5. In a nutshell I Consider the (fictious) product group alcoholic beverage with bubbles I Crisis basket : 15 / 19

42 5. In a nutshell I Consider the (fictious) product group alcoholic beverage with bubbles I Crisis basket : 15 / 19

43 5. In a nutshell I Consider the (fictious) product group alcoholic beverage with bubbles I Crisis basket : 15 / 19

44 5. In a nutshell I Consider the (fictious) product group alcoholic beverage with bubbles I Crisis basket : 15 / 19

45 5. In a nutshell Consider the (fictious) product group alcoholic beverage with bubbles Crisis basket (option B) : 16 / 19

46 5. In a nutshell I Consider the (fictious) product group alcoholic beverage with bubbles I Crisis basket (option B) : 16 / 19

47 5. In a nutshell I Consider the (fictious) product group alcoholic beverage with bubbles I Crisis basket (option B) : 16 / 19

48 5. In a nutshell I Consider the (fictious) product group alcoholic beverage with bubbles I Crisis basket (option B) : 16 / 19

49 5. In a nutshell The model success in one graph Figure 12. Figure Model7: Estimated Actual and Actual estimated Within growth Components on the of share Expenditure of foreign Switching goods in expenditures Y o y change in import expenditure share Data CES Non homothetic / 19

50 6. Where to go now? It is inter temporal wealth more than income that matters fro demand : asset prices went down future income growth was revised downwards Permanent income might have fallen more than current income, which would reinforce the expenditure switch. The credit crunch in from 2008 to 2011 has also increased the shadow value of one spent lats or decreased income deflated by the full value of a lats. Commentators also mention an increases in perceived uncertainty (???). If so, this also reduces the share of income devoted to consumption. Those different effects call for a permanent income model in which credit constraints and expectations would have an important quantitative role. 18 / 19

51 6. Where to go now? It is inter temporal wealth more than income that matters fro demand : asset prices went down future income growth was revised downwards Permanent income might have fallen more than current income, which would reinforce the expenditure switch. The credit crunch in from 2008 to 2011 has also increased the shadow value of one spent lats or decreased income deflated by the full value of a lats. Commentators also mention an increases in perceived uncertainty (???). If so, this also reduces the share of income devoted to consumption. Those different effects call for a permanent income model in which credit constraints and expectations would have an important quantitative role. 18 / 19

52 6. Where to go now? It is inter temporal wealth more than income that matters fro demand : asset prices went down future income growth was revised downwards Permanent income might have fallen more than current income, which would reinforce the expenditure switch. The credit crunch in from 2008 to 2011 has also increased the shadow value of one spent lats or decreased income deflated by the full value of a lats. Commentators also mention an increases in perceived uncertainty (???). If so, this also reduces the share of income devoted to consumption. Those different effects call for a permanent income model in which credit constraints and expectations would have an important quantitative role. 18 / 19

53 6. Where to go now? It is inter temporal wealth more than income that matters fro demand : asset prices went down future income growth was revised downwards Permanent income might have fallen more than current income, which would reinforce the expenditure switch. The credit crunch in from 2008 to 2011 has also increased the shadow value of one spent lats or decreased income deflated by the full value of a lats. Commentators also mention an increases in perceived uncertainty (???). If so, this also reduces the share of income devoted to consumption. Those different effects call for a permanent income model in which credit constraints and expectations would have an important quantitative role. 18 / 19

54 6. Where to go now? It is inter temporal wealth more than income that matters fro demand : asset prices went down future income growth was revised downwards Permanent income might have fallen more than current income, which would reinforce the expenditure switch. The credit crunch in from 2008 to 2011 has also increased the shadow value of one spent lats or decreased income deflated by the full value of a lats. Commentators also mention an increases in perceived uncertainty (???). If so, this also reduces the share of income devoted to consumption. Those different effects call for a permanent income model in which credit constraints and expectations would have an important quantitative role. 18 / 19

55 6. Where to go now? It is inter temporal wealth more than income that matters fro demand : asset prices went down future income growth was revised downwards Permanent income might have fallen more than current income, which would reinforce the expenditure switch. The credit crunch in from 2008 to 2011 has also increased the shadow value of one spent lats or decreased income deflated by the full value of a lats. Commentators also mention an increases in perceived uncertainty (???). If so, this also reduces the share of income devoted to consumption. Those different effects call for a permanent income model in which credit constraints and expectations would have an important quantitative role. 18 / 19

56 6. Where to go now? It is inter temporal wealth more than income that matters fro demand : asset prices went down future income growth was revised downwards Permanent income might have fallen more than current income, which would reinforce the expenditure switch. The credit crunch in from 2008 to 2011 has also increased the shadow value of one spent lats or decreased income deflated by the full value of a lats. Commentators also mention an increases in perceived uncertainty (???). If so, this also reduces the share of income devoted to consumption. Those different effects call for a permanent income model in which credit constraints and expectations would have an important quantitative role. 18 / 19

57 19 / 19