Adjustment Mechanisms in a Currency Area

Abstract

Both the euro-area and the United States suffered an initially quite similar housing and financial shock in 2007/8, with several states in both regions being particularly badly affected. Yet there was never any question that the worst hit US states would need a special bail-out or leave the dollar area, whereas such concerns have worsened in the euro-area. We focus on three badly affected states, Arizona, Spain and Latvia, to examine the working of relative adjustment mechanisms within the currency region. We concentrate on four such mechanisms, relative wage adjustment, migration, net fiscal flows and bank flows. Only in Latvia was there any relative wage adjustment. Intra-EU migration has increased, but is more costly for those involved in the EU (than in the USA). Net federal financing helped Arizona and Latvia in the crisis, but not Spain. The locally focussed structure of banking amplified the crisis in Spain, whereas the role of out-of-state banks eased adjustment in Arizona and Latvia. The latter reinforces the case for an EU banking union.

Appendices

Appendix 1: Models and Regression Results

1.1 Labour Markets: Wage Adjustment

1.1.1 Δ Relative RULC_t = a + b₁ Relative UE_t + b₂ Δ Relative UE_t

Δ Relative RULC_t :

Change in relative real unit labour cost (RULC). Percentage change in the ratio of state RULC to federal RULC (USA for Arizona and EU for Spain and Latvia) at time t.

Relative UE_t :

Relative unemployment. Ratio of state unemployment rate to federal unemployment rate (USA for Arizona and Euro area for Spain and Latvia) at time t.

Δ Relative UE_t :

Change in relative unemployment over four quarters at time t.

[Example] Let’s look at Latvia’s Δ Relative UE_t=2Q09

$$ \varDelta\;\mathrm{Relative}\ \mathrm{U}{{\mathrm{E}}_{{\mathrm{t}=2\mathrm{Q}09}}}=\mathrm{Relative}\ \mathrm{U}{{\mathrm{E}}_{{\mathrm{t}=2\mathrm{Q}09}}}-\mathrm{Relative}\;\mathrm{U}{{\mathrm{E}}_{{\mathrm{t}=2\mathrm{Q}08}}}=1.85\text{--} 0.88=0.97 $$

Arizona

• Dependent Variable: RULC_AZ_PC

• Sample (adjusted): 2001Q4 2011Q3

• Included observations: 40 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  UE_AZ	−0.587491	1.583490	−0.371011	0.7127
  UE_AZ_4	0.631488	1.538440	0.410473	0.6838
  C	0.553782	1.542008	0.359131	0.7215
  R-squared	0.005356
  Adjusted R-squared	−0.048409

Spain

• Dependent Variable: RULC_ES_PC

• Sample (adjusted): 2001Q4 2011Q3

• Included observations: 40 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  UE_ES	−0.209835	0.367222	−0.571412	0.5712
  UE_ES_4	1.959192	0.717786	2.729492	0.0097
  C	0.189918	0.504838	0.376197	0.7089
  R-squared	0.198689
  Adjusted R-squared	0.155375

Latvia

• Dependent Variable: RULC_LV_PC

• Sample (adjusted): 2001Q4 2011Q3

• Included observations: 40 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  UE_LV	−2.782770	0.701567	−3.966505	0.0003
  UE_LV_4	−1.155908	0.813785	−1.420410	0.1639
  C	4.904885	0.966668	5.074012	0.0000
  R-squared	0.437225
  Adjusted R-squared	0.406805

1.2 Labour Markets: Migration

1.2.1 Mig_t = a + b₁ Relative UE_t

Mig_t :

Net migration. Net migration as percentage of state population at time t.

Relative UE_t :

Relative unemployment rate. Ratio of state unemployment rate to federal unemployment rate (USA for Arizona and Euro area for Spain and Latvia) at time t.

Arizona

• Dependent Variable: MIG_AZ

• Sample (adjusted): 2000 2011

• Included observations: 12 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  R_UE_AZ	−6.878790	2.090633	−3.290291	0.0081
  C	8.209843	2.061695	3.982084	0.0026
  R-squared	0.519831
  Adjusted R-squared	0.471814

Spain

• Dependent Variable: MIG_ES

• Sample (adjusted): 2002 2010

• Included observations: 9 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  R_UE_ES	−1.505587	0.112083	−13.43274	0.0000
  C	3.205591	0.157877	20.30441	0.0000
  R-squared	0.962654
  Adjusted R-squared	0.957319

Latvia (census)

• Dependent Variable: MIG_LV_CEN

• Method: Least Squares

• Date: 10/09/12 Time: 16:58

• Sample: 1998 2011

• Included observations: 14

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  R_UE_LV	−0.789801	0.302457	−2.611280	0.0227
  C	0.428841	0.432768	0.990927	0.3413
  R-squared	0.362339
  Adjusted R-squared	0.309201

Latvia (OECD-based)

• Dependent Variable: MIG_LV_OECD

• Sample (adjusted): 2000 2010

• Included observations: 11 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  R_UE_LV	−1.450009	0.580998	−2.495722	0.0341
  C	1.180388	0.820604	1.438438	0.1842
  R-squared	0.409008
  Adjusted R-squared	0.343342

1.2.2 Mig_t = a + b₁ Relative UE_t + b₂ Mig_t−1

Latvia (census)

• Dependent Variable: MIG_LV_CEN

• Sample (adjusted): 1999 2011

• Included observations: 13 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  R_UE_LV	−0.676919	0.226290	−2.991378	0.0135
  MIG_LV_CEN(−1)	0.699093	0.211795	3.300799	0.0080
  C	0.668269	0.330433	2.022401	0.0707
  R-squared	0.700140
  Adjusted R-squared	0.640168

Latvia (OECD-based)

• Dependent Variable: MIG_LV_OECD

• Sample (adjusted): 2001 2010

• Included observations: 10 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  R_UE_LV	−0.892004	0.376736	−2.367718	0.0498
  MIG_LV_OECD(−1)	1.131123	0.279578	4.045821	0.0049
  C	0.993964	0.464170	2.141380	0.0695
  R-squared	0.854781
  Adjusted R-squared	0.813290

1.3 Fiscal Differences

1.3.1 Tran_t = a + b₁ UE_t + b₂ UE_t−1

Trant:

Federal net transfer. Federal net transfer as % of state GDP at time t.

Relative UE_t :

Relative unemployment rate. Ratio of state unemployment rate to federal unemployment rate (USA for Arizona and Euro area for Spain and Latvia) at time t.

Arizona

• Dependent Variable: TRAN_AZ

• Sample (adjusted): 1992 2010

• Included observations: 19 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  UE_AZ	10.52150	14.21886	0.739968	0.4700
  UE_AZ(−1)	−1.186225	14.33070	−0.082775	0.9351
  C	−5.302360	16.02802	−0.330818	0.7451
  R-squared	0.034813
  Adjusted R-squared	−0.085835

Spain

• Dependent Variable: TRAN_ES

• Sample (adjusted): 2001 2010

• Included observations: 10 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  UE_ES	−1.409257	0.582521	−2.419241	0.0461
  UE_ES(−1)	1.269782	0.756681	1.678094	0.1372
  C	1.003890	0.532544	1.885086	0.1014
  R-squared	0.476483
  Adjusted R-squared	0.326906

Latvia

• Dependent Variable: TRAN_LV

• Sample (adjusted): 2001 2010

• Included observations: 10 after adjustments

  Variable	Coefficient	Std. Error	t-Statistic	Prob.
  UE_LV	1.220392	1.107322	1.102111	0.3069
  UE_LV(−1)	−0.736916	1.325207	−0.556076	0.5955
  C	1.154554	1.384201	0.834094	0.4318
  R-squared	0.152870
  Adjusted R-squared	−0.089167

Appendix 2: Data Tables

Table 1 Unemployment rate. % of labour force, seasonally adjusted

Table 2 Standard deviation of unemployment rate. Using population of 2010

Table 3 Unemployment cycle in US states. Seasonally adjusted

Table 4 Unemployment cycle in European states. Seasonally adjusted

Table 5 Standard deviation of housing price index. Index 2006 = 100

Table 6 Real labour unit cost (RULC). Index 3Q08 = 100 unless noted otherwise

Table 7 Net migration flows. % of total population

Table 8 Federal net transfer to state. % of state GDP

Table 9 Size of domestic banking sector in deposit market. % of deposits

Table 10 Size of domestic banking sector in lending market. % of loans