Exchange Rate Volatility and Employment Growth: Empirical Evidence from the CEE Economies

1
Exchange Rate Volatility and Employment Growth
Empirical Evidence from the CEE Economies

• byAnsgar Belke and Ralph Setzer
• University of Hohenheim (Germany)

Paper prepared for the EABCN Workshop on Business
Cycles and Acceding Countries Vienna, 23-24 April
2004
2
Main interest of the paper
Impact of Exchange Rate (EXR) uncertainty on job
creation and destruction flows
Earlier Studies EXR volatility has negative
effects for labor markets

• Intra-European EXR variability (Belke, Gros,
  2001)
• Transatlantic EXR variability (Belke, Gros,
  2002a)
• Mercosur EXR variability (Belke, Gros, 2002)

? same results for the CEE countries?
? debate on the costs and benefits of an early
adoption of the euro compared with those of
introducing the euro at a later stage
3
Outline of the paper
Chapter 1 status-quo ante in terms of CEEC trade
integration with the eurozone and in terms of the
discussion on early euroization

• Chapter 2 simple model of investment and
  uncertainty explaining the transmission channel
  of the negative relation between uncertainty and
  employment

Chapter 3 data and definitions
Chapter 4 presents and comments the regression
results
Chapter 5concludes and derives some policy
conclusions
4
Ch. 1 CEECs The status-quo ante

• Date of entry to EU May 2004
• ? Earliest date of entry to Eurozone 2007

• 2 objections against early adoption of the euro
• Converge first and durably, then join
• Forced exit from EXR pegs (Czech Republic, Poland)

• Status quo
• High trade integration of CEECs with present EU
  countries

• Similar labor market rigidities as EU countries
  (Riboud et al. 2002)

• Functioning currency-union in their neighborhood

• Efforts to fulfil Maastricht criteria

5
Why should policymakers care about exchange rate
volatility?

• R. Mundell (2000) Threat to Prosperity, in Wall
  Street Journal Europe, March 30th
• It is time to end benign neglect of exchange
  rates

• R. Mundell (2000a) AER, Vol. 80
• The volatility of exchange rates is especially
  disturbing among countries each of which have
  achieved ... price stability. The volatility
  therefore measures real-exchange-rate changes and
  involves dysfunctional shifting between domestic
  and international goods industries and aggravates
  instability in the financial markets (p. 338)

6
Volatility particularly costly for emerging
market economies

• EXR volatility with negative effects on
  investment and economic growth in developing
  countries (Calvo and Reinhart, 2000a Reinhart
  and Reinhart, 2001).

• Capital markets in Emerging Markets are of an
  incomplete nature (Dornbusch 2001)

? no possibility for hedging the EXR risk
? higher interest rate premium
? (Irrevocably) fixed exchange rates contribute
to lower cost of capital and thereby higher growth
7
Higher pass-through from EXR swings to inflation
due to higher openness of these countries (Chang
and Velasco 2000)
1. direct channel in that the EXR affects
domestic currency prices of imported goods,
2. indirect channel through variations in
relative prices between domestic and foreign
goods which in turn affect domestic aggregate
demand, and inflation.

• Less exchange rate volatility leads to more
  international trade (Frankel and Rose 2002)
• Trade within a currency union is two or three
  times larger than a gravity model of
  international trade would suggest

8
Option value of waiting

• Even short-term spikes of volatility can have a
  strong negative impact on investment and job
  creation due to set-up costs

• Market access costs

• Capital costs

• Hiring and firing costs

Result Impact on investment and employment
possible, even if trade quantities react little
in the short term.
9
Ch. 2 The model

• Model in the tradition of Dixit 1989 and
  Pissarides 2000 Uncertainty of future earnings
  raises the option value of waiting with decisions
  with concern investment projects in general

• Here Extension by modeling the labor market
  explicitly (interpretation job creation instead
  of investment)

• To create a job, one needs to sustain a sunk cost
  (hiring, training, provision of job-specific
  capital)

? Even a temporary short-run increase in
uncertainty can have a strong and lasting impact
on the unemployment rate
10
Model ingredients

• 3 periods model, no discounting, risk neutrality

• unit start-up costs in job creation period (0 or
  1)

• output of the worker is sold in the following
  period in a foreign market at domestic price p
  with a certain component p (the foreign price)
  plus a stochastic component e (the exchange rate)

? return to investment in t1 (if investment in
t0) and t2
11
Main elements driving the results

• wage bargaining process wage rate w for the job
  is determined by (generalized) Nash bargaining
  solution that maximizes a weighted product of the
  workers and the firms expected net return from
  the job

• basic scenario implicit assumption that firm and
  worker sign binding employment contract for two
  periods (zero and one). ? no possibility of job
  termination in period 1 whenever the exchange
  rate turns out to be unfavorable (realistic for
  short period length)

• systematic comparison of (unconditional) expected
  returns of job creation today or tomorrow

12
Bargaining problem for a job created in period 0
13
First element Probability that it will not be
worthwhile to open a job
Second element Product of the probability that
it is worthwhile to open a job and the expected
value of the net return to the firm under this
outcome
14
Results

• ? A firm prefers to wait if and only if
• (5)

15
Note
is

• increasing in (expected return of a filled job
  per period decreasing in fallback wage ),

16
Conclusions

• adverse impact of EXR uncertainty on job creation
  and employment should be stronger if labor market
  is characterized by generous unemployment benefit
  systems, powerful trade unions, minimum wage
  restrictions or large hiring costs (interactions!)

• However, even if there are no firing costs and if
  workers can be laid off at any point in time,
  exchange rate uncertainty should have a direct
  impact on job destruction (Scenario B).

? even short-term spikes of uncertainty may have
strong impact on job creation
17
Conclusions (2)

• 'option value of waiting' increasing in
• intuition the higher the variance of e, the
  higher the potential losses the firm can avoid by
  delaying job creation

• EXR uncertainty can have a negative impact not
  only on job creation but also on job destruction
  flows

? impact seems to be less ambiguous than in
hysteresis models where the main conclusion is
EXR uncertainty delays hiring and firing and
elimination fosters structural change
18
Does the model apply to the CEEC labor markets?

• Impact of EXR uncertainty on job destruction
  strong if labor market is characterized by high
  rigidities.

• Countries range somewhere in the middle of the
  flexibility scale compared to the EU economies
  (Riboud et al., 2002).

• CEECs with high payroll and other taxes as well
  as strong employment protection legislation

• CEECs are required to align their legislation
  with the acquis communautaire which includes a
  number of provisions regarding the labor market
  regulations

? CEECs with similar rigidities that are
troubling the EU countries (Belke and Hebler,
2001, 2002)
19
Table 1 Labour market flexibility in the CEECs
How large are the costs of job creation and the
fallback wage?
Numbers in brackets refer to the new labour
code if approved EU average without
Luxembourg and Greece 1 minimum protection,
6 maximum protection Weighted average of
the first three column
Source Riboud et al. (2001)
20
Ch. 3 Data and Definitions

• Which exchange rates?
• 10 times 30 volatilities of the nominal bilateral
  euro EXR,

• 10 times 30 volatilities of the real bilateral
  euro EXR,

• 10 effective volatilities of the nominal EXR
  (weighted bilateral volatilities)

• 10 effective volatilities of the real EXR
  (weighted bilateral volatilities)

21

• Variability as standard deviation of monthly
  percentage changes (1991-2001)

• Monthly instead of more short-term exchange rate
  data (longer time horizon of investors) ?
  disadvantage annual data

• Nominal exchange rates, since nominal and real
  EXR highly correlated on a monthly basis

• Actual variability (instead of option prices or
  unanticipated ones), since at monthly horizon
  anticipated change of EXR negligible

22

• Panel of 10 CEECs
• Bulgaria (BG), Czech Republic (CZ), Estonia
  (EE), Hungary (HU), Latvia (LV), Lithuania (LT),
  Poland (PL), Romania (RO), Slovak Republic (SK),
  Slovenia (SL)

• Sample ranges from 1992-2001

• Real GDP growth as cyclical control

• Panel unit root tests (Levin and Lin 1992),
  differencing variables till stationarity

• Focus on Employment data (in line with our model)

23
Ch. 4. Empirical Analysis and Results
24
(No Transcript)
25

• ? Tables 2a and 2b reveal overwhelming evidence
  in favor of a fixed effects estimation

26

• Test for dynamic effects in order to capture
  speed of adjustment of labor markets ? include
  lagged employment variables as regressor.

27

• Estimating first-order model substantial
  complications due to heterogeneity of
  cross-sections

• Main problem to be treated correlation of lagged
  dependent variable (level of employment) with
  disturbance, even if the latter does not exhibit
  autocorrelation itself.

• While taking first differences enables one to get
  rid of heterogeneity, i.e., the group effects,
  the problem of correlation between the lagged
  dependent variable and the disturbance still
  remains. Moreover, a moving-average error term
  now appears in the specification.

28

• However, the treatment of the resulting model is
  a standard application of the IV approach. The
  transformed model looks as follows

29
Estimation procedure

• FGLS estimates of a model assuming
  cross-sectional heteroscedasticity and
  autocorrelation but without correction for
  contemporaneous correlation (assumption shocks
  to labor markets asymmetric) (Tab. 3).

• SUR estimates of a model with heteroscedasticity
  and contemporaneous cross-sectional correlation
  (assumption symmetric shocks) (Tab. 4 and 5).

• In some cases AR error term to get rid of
  autocorrelation problems in time dimension with
  restriction of a common autocorrelation
  coefficient across countries.

30
Summary of Results

• Results rather strong in many cases EXRV
  significant impact on employment growth.

• Data confirm that economies with closer ties with
  the euro zone (e.g., Czech Republic) show
  stronger impact of euro EXRV.

• Systematic correlation between openness and the
  strength of impact of EXRV on employment growth
  corresponds to general finding that for emerging
  markets this channel is important.

31
Robustness Checks

• First check Limit sample to a group of rather
  homogenous CEECs with respect to labor market
  regulation, namely the Visegrád countries.

• Second check Including indicators of strictness
  of employment protection legislation.

• Third check Implement real wage growth in order
  to check whether significant relationship between
  EXV and employment growth driven by missing third
  variable related to labor costs.

• Exogeneity of volatility and robustness variables
  with respect to the change of the employment rate
  checked by extensive Granger causality tests.

32
Ch. 5 Conclusion

• EXR variability has a significant negative effect
  on employment growth in the CEECs.

• Correlation between openness and the strength of
  the impact of EXR volatility on labor markets.

• It was argued that these results are due to the
  degree of irreversibility of all employment
  decision.

• Results confirm earlier studies to EMS,
  transatlantic and Mercosur ER variability

33
Timing of Euro adoption

• In developing countries co-movement of exchange
  and interest rates variability?

• Reducing EXR variability could bring substantial
  benefits (Dornbusch 2001 Frankel and Rose 2002)

• Early adoption of the euro reduces the degree of
  uncertainty and has the same effects as the
  removal of employment-protection legislation and
  other restrictions of hiring and firing

• Underlying policies must be compatible with this
  choice