Equilibrium ER • PPP: – Empirical evidence: • Define: – Test the behavior of RER » If RER is non-stationary -> PPP does not hold » If RER is stationary -> PPP might hold » Conclusion (Rogoff (1986)): RER is mean reverting albeit very slowly with half statistics of mean reversion around 4 years. This is too long time period to be explained by reasonable nominal rigidities or barriers to arbitrage. 1 * p e p p e p RER * Rogoff, K. (1996),”The Purchasing Power Parity Puzzle”, Journal of Economic Literature, vol. 34, June 1996, pp.:647-668
43
Embed
Equilibrium ER PPP: –Empirical evidence: Define: –Test the behavior of RER »If RER is non-stationary -> PPP does not hold »If RER is stationary -> PPP.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Equilibrium ER
• PPP: – Empirical evidence:
• Define:– Test the behavior of RER
» If RER is non-stationary -> PPP does not hold» If RER is stationary -> PPP might hold» Conclusion (Rogoff (1986)): RER is mean reverting
albeit very slowly with half statistics of mean reversion around 4 years.
This is too long time period to be explained by reasonable nominal rigidities or barriers to arbitrage.
1*
p
ep
p
epRER
*
Rogoff, K. (1996),”The Purchasing Power Parity Puzzle”, Journal of Economic Literature, vol. 34, June 1996, pp.:647-668
Fact: RER is highly persistent• Implications:
– Problems with econometric inference– We need to explain these large swings in RER
• Unfortunately, many theories are lacking micro-foundations• Possibly the best explanations – Balassa-Samuelson efect:
– Poorer countries have cheaper price level– Improvement in productivity in tradable sector (relative to foreign
country) appreciate ER.– Improvement in productivity in non-tradable sector (relative to foreign
country) depreciate ER
Balassa, B. (1964),”The Purchasing Power Parity Doctrine: A Reappraisal”, Journal of Political Economy, vol. 72, December, pp. 584-596
Samuelson, P. (1964),”Theoretical Notes on Trade Problems”, Review of Economics and Statistics, vol. 23, pp.: 1-60
Harrod, R. F. (1933), International Economics, London: James Nisbet and Cambridge University Press
Empirical evidence : B-S effect
y = 0.78x - 3.66
R2 = 0.86
-1.5
-1.0
-0.5
0.0
0.5
3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0 5.2
Logarithm of price level
Logarithm of GDP per capita (in PPP terms) Richer country
More expensive country
B-S is not enough– MacDonald and Ricci (2001): Enlarged standard B-S
framework by distribution sector.– MacDonald and Ricci (2002): Tested implication of new
trade theory on ER. (They focus on imperfect substituability of tradables and on the importance of competitiveness)
– others
• There is no theory (with proper micro-foundations) satisfactory explaining behavior of RER
MacDonald, R. and Ricci, L. (2001),”PPP and the Balassa Samuelson Effect: The Role of the Distribution Sector”, IMF Working Paper, WP/01/38
MacDonald, R. and Ricci, L. (2002),”Purchasing Power Parity and New Trade Theory”, IMF Working Paper, WP/02/32
Example of other theories
• Monetary model of ER:
iymp 21
*2
*1
** iymp *pep
)()( *2
*1
* iiyymme
• Let PPP condition hold only in the long-run, and assume prices are sticky in the short-run -> Dornbusch overshooting model.
• Various approaches– Most common: BEER and FEER– Persistent behavior of RER -> problems with
econometric estimates– Assumption of non-stationarity cannot be
satisfactory rejected due to ‘short sample’– In all BEER estimations, RER is assumed to be
I(1) – i.e. simple PPP does not hold
BEER
• Most common
• Tries to find direct behavioral link between RER and fundamentals (assumes variables to be I(1)).
• Absolutely no micro foundation
• Fundamentals: everything one suspect might affect RER
REER indices
65
70
75
80
85
90
95
100
105
93 94 95 96 97 98 99 00 01 02 03 04
REER CPI based REER PPI based REER PPI (SK manufacturing) based
appreciate
RER (de-trended)
-15%
-10%
-5%
0%
5%
10%
15%
20%
93 94 95 96 97 98 99 00 01 02 03 04
REER CPI based REER PPI based
REER PPI (SK manufacturing) based REER ULC based
overvaluation
undervaluation
In-sample approach – estimates are likely biased
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
94 95 96 97 98 99 00 01 02 03
example of RER 'true equilibrium' of RER
estimated BEER
-50%
-40%
-30%
-20%
-10%
0%
10%
20%
30%
94
95
96
97
98
99
00
01
02
03
'true deviations from equilibrium
estimated exchange rate misalignment
Unit root tests
• Single time serries– We performed: ADF , KPSS Kwiatkowski-
Phillips-Schmidt-Shin (1992) , Ng-Perron Ng-
Perron (2001) – Conclusion: All series are I(1)
Kwiatkowski, D., Phillips, P. C. B., Schmidt, P. and Shin, Y. (1992),”Testing the Null Hypothesis of Stationary against the Alternative of a Unit Root”, Journal of Econometrics, vol. 54, pp.: 159-178
Ng, S. and Perron, P. (2001),”Lag Length Selection and the Construction of Unit Root Tests with Good Size and Power”, Econometrica, vol. 69, pp.: 1519-1554
Fundamentals• We chose:
» Proxy for Ballassa-Samuelson effect (e.g. relative price of non-tradables vs tradables (NT) or relative labour productivity in tradables vs nontradables (LP))
» (ToT) Terms of trade (i.e. export prices relative to import prices) intend to capture changes in international economic environment.
» (gov) Share of government spending on GDP - intends to capture the effects of fiscal policy.
» (rr) Real interest rate differential. Higher interest rate differential should attract more capital inflows.
» Proxy for country risk. We used following proxies: country rating (rat), spreads on Eurobonds (spread), our currency vulnerability index adjusted for movements in RER (vul); NFA
Est
imat
ion
of B
EE
R
Single country approach
Able to estimate country specific behavior
Short sample (about a decade) , RER is highly persistent Estimates are not robust
RER is likely to be undervalued for longer period in the first phase of transition
Estimates are biased
Plenty of ad hoc-factors affecting early stage of transition
Possibility of structural breaks
Questionable quality of data on the early stage of transition
Measurement error
Cross section approach Missing time series dimension
(Impossible to account for country specific effects)
Panel data approach
In-sample estimates
Out-of-sample estimates
Including only accession countries
Partially helps to solve problems connected with short sample
Estimates are still biased
Using data from early stage is still questionable
Including broader set of countries
Heterogeneity ?
Helps to solve problems connected with short sample
Biased L-R estimates
Biased constant term
Heterogeneity ?
Estimates are not biased finally
No constant for accession country (country-specific effect)Heterogeneity ? (is homogenous long run behavioral assumption for each country plausible?)
+
-
-
-
+
-
+
+
-
Aims to estimate ‘specific’ behavior of countries in transition
• Other country specific conditions affecting L-R:– Fixed effects
Panel Unit root tests
• We performed:– Levin, Lin and Chu t, Levin, Lin and Chu (2002);
Breitung t-stat, Breitung (2000)
– Im, Pesaran and Shin W-stat, Im, Pesaran and Shin
(2003); ADF - Fisher Chi-square, PP - Fisher Chi-square, Maddala and Wu (1999) and Choi(2001)
– Hadri Z-stat, Hadri (2000)
Conclusion: All variables are I(1) rather than I(2) or I(0)
ReferencesLevin, A., Lin, C. F., and Chu, C. (2002),”Unit Root Tests in Panel Data: Asymptotic
and Finite-Sample Properties,” Journal of Econometrics, Vol. 108, pp.:1-24
Breitung, J. (2000),”The Local Power of Some Unit Root Tests for Panel Data,” in Baltagi (ed.), Advances in Econometrics, Vol. 15: Nonstationary Panels, Panel Cointegration, and Dynamic Panels, Amsterdam: JAI Press, pp.:161-178
Im, K. S., Pesaran, M. H., and Shin, Y. (2003),”Testing for Unit Roots in Heterogeneous Panels”, Journal of Econometrics, Vol. 115, pp.: 53-74
Maddala, G. S., and Wu, S. (1999),”A Comparative Study of Unit Root Tests with Panel Data and A New Simple Test, Oxford Bulletin of Economics and Statistics, Vol. 61, pp.:631-652
Choi, I. (2001),”Unit Root Tests for Panel Data,” Journal of International Money and Finance, Vol. 20, pp.:249-272
Hadri, K. (2000).”Testing for Stationarity in Heterogeneous Panel Data,” Econometrics Journal, Vol. 3, pp.:148-161
Cointegration tests
• We used following residual based cointegration tests:– Kao (1997) – Kao (1999) – Pedroni (1995) – Pedroni (1999)
(15 tests together)
critical probability critical probability
Residual based panel cointegration tests
4 explanatory variables (excluding ToT)
3 explanatory variables (excluding ToT and fiscal
spending)
0.70%
Kao (1997)
DF_Rho -5.1 0.00% -2.5
7.40%
DF_Rho* -9.5 0.00% -6.4 0.00%
DF_t_Rho -2.8 0.20% -1.4
3.60% DF_t_Rho* -3 0.20% -1.8
Kao (1999)
7.30%ADF -2.7 0.40% -1.5
Pedroni (1995)
0.00%
TN1_rho -15.4 0.00% -10.9 0.00%
t_rho_NT -75.8 0.00% -69.1
0.00% TN2_rho -15.2 0.00% -10.7
Pedroni (1999)
20.00%Standarized Panel V-statistics -0.7 -0.8
Panel V-statistics 18
24.80%
26.2
11.80%Standarized Panel Rho-statistics -7.8 1.2
Panel Rho-statistics -93.6
0.00%
-22.5
19.50%Standarized Panel PP-statistics -7 0.9
Panel t-statistics (nonparametric) -11.2
0.00%
-5.1
0.00%Standarized Panel ADF-statistics -504 -267
Panel t-statistics (parametric) -489
0.00%
-251
Group Rho-statistics -93.8
0.00%
-22.9
1.90%
3 explanatory variables (excluded ToT and fiscal spending)
• Koruna will definitely be appreciating in real terms, at least for the next two decades. This is connected with fact that the country is economically just at 40% of EU average and this huge gap will continue to be diminishing slowly in the future.
• It is tricky to estimate the precise speed of sustainable appreciation. Data problems, short sample, and structural changes are the obvious challenges. First results point to more than 1.7% real appreciation in the years ahead. Then the pace of appreciation is likely to slow down to about 0.5-1.5% in the medium- to longer-term.
• As for the current equilibrium level of Koruna, it is even more tricky to estimate it. According to the preliminary research, koruna is slightly undervalued at present – equilibrium being about Sk39.3/Eur (June). Our estimates comfortably support Sk39/€ in 4Q04 to be fully in line with macro picture or even slightly over-valuated. Furthermore, based on the preliminary (!) data, Koruna could be entering ERMII in 2006 at Sk36-37/€.
• Worries of CB about too strong ER at the moment might not be justified