The Long-Run Determinants of the Real Exchange Rate in T

Document Sample
The Long-Run Determinants of the Real Exchange Rate in T Powered By Docstoc
					The Long-Run Determinants
of the Real Exchange Rate in
Transition Economies
MSc: CĂLĂVIE ALINA MIHAELA
Supervizor: ALTĂR MOISĂ
CONTENT
 Literature Review
 Theoretical Background
 Determinants of the Real Exchange
 Rate in Transition Economies: A Model
 Empirical Analysis for Romania
 Conclusion
       1. LITERATURE REVIEW

Fundamental                Behavioural
Models                     Models

Williamson (1983) - FEER   Clarck & MacDonald (1998) - BEER
Williamson (1994)          Stain & Allen (1995) - NATREX
                           Feyzioglu (1997)
                           Elbadawi (1995)
                           Halpern & Wyplosz (1997, 2002)
                           Frait & Komarek (2001)
2. THEORETICAL BACKGROUND
  Real Exchange Rate and External
  Competitiveness
Real appreciation can be interpreted as a loss of
  competitiveness only if the RER becomes overvalued
  in relation to its equilibrium value.

The competitiveness must be interpreted with respect
  to the main trading partners, different world regions
  and different group of producers.
  Real Exchange Rate, Double-Speed
  Economy and Deindustrialization

Due to the weak institutional framework, the inefficient
  domestic firms are not forced to leave the market
  and they burden the cost of the efficient ones.

Keeping the trained employees in the old sector may
  artificially delay the restructuring process.

Deindustrialization and real appreciation are
  simultaneously determined by the productivity gains
  in industrial production.
   Real Exchange Rate as an Indicator of
   Convergence
Real exchange rate development can be seen as a common
  denominator since the external purchasing parity of transitional
  countries’ currencies is determined primarily by their relative
  productivities. The real appreciation also implies the convergence
  of price levels of these countries to the EU countries.

                          R    Appreciation
                      (ERDI)

                                   ERER (successful transformation)


                                                ERER (unsuccessful transformation)
                 ERDI>1
                               Initial real
                               undervaluation
                 R=ERDI=1
                                                                      PPP
                 ERDI<1

                                                                      Time
What is the equilibrium real exchange rate?

  PPP – Purchasing Power Parity
It assumes that the real exchange rate equilibrium remains
   stable over a long period. It does not take into account for
   structural changes.

  ERER – Equilibrium Real Exchange Rate:
  the rate consistent with the country’s macroeconomic
  balance, including both internal and external balances.
Any change in fundamental determinants of internal or external
  balance can affect the value of the ERER.
  DETERMINATS OF THE RER IN A
SMALL OPEN TRANSITIONAL ECONOMY

    Definition of the RER
 The real exchange rate can be defined as the nominal exchange
   rate (E ), times the ratio of foreign to domestic prices (P*/P).

                           RER  EP * / P
 where the domestic price index is:
                                             (1 a b )
                        P  P P Pn
                                    a   b
                                        m   x
 and the foreign price index is:

                       P*  Pna*  Pxb*  Pm1a* b*)
                              *      *
                                            (
                                              *
                  THE MODEL

    Hypotheses
-   the currency convertibility is already established
-   there are no quantitative restrictions on foreign trade
-   managed float exchange rate regime
-   the prices of tradables and nontradables are liberalized
-   strong monopolistic structure
-   subsidization of the nontradables sector
-   price rigidity and inflexible wages
 Consumer’s optimization problem:
          
          
                               i  i     i
                     max U i Cn , Cm , M d 
           pn  C n  pm  C m  M d  V  M 0
                   i          i       i       i



Profit maximization in the nontradables sector:
              pn  (1  j )[ wLn / Qn  sn ]

                pn  pn ( j, w, sn , C )

Profit maximization in the tradables sectors:
                    pi  pi ( pi* , E )

           Li  Li ( w, pi , si )
                                          i  m, x
          Qi  Qi ( w, pi , si )
   Given managed exchange rates:
                                    CA  KA  F1  F0
   Total product, in nominal terms, equals:
                   Y  ( pn  sn )  Qn  pm  Qm  ( p x  s x )  Qx
   The budget deficit can be written, in nominal terms, as:
D  pn  Gn  (1 / E )  pm  Gm  sn  Qn  s x  Qx  (1 / E )  x  p * Q x (1 / E )  f  pm  I m
                          *
                                                                         x
                                                                                                 *


   The overall change in the money supply is given by:
                                 M s1  M s 0  D  ( F1  F0 )
   The nominal exchange rate is market determined, so:
                               E  E[ F , D(G, s, taxes ), w, i]
   Consequently, the real exchange rate can be written as:

RER  RER [ pn ( j , w, sn , C n ), pm ( pm , E ), p x ( p * , E ), D(G, s, taxes ), F , w, i ]
                                          *
                                                           x
 A Basic Introduction to the Romanian
           Economic Context
   Price liberalization
The initial process was built up in three rounds: in November 1990,
   in April 1991 and in July 1991
During 1994-1996 controls were reintroduced, especially on food
   prices.
In early 1997 most prices were liberalized. The weight of
   administrated prices in the CPI basket reduced at about 15
   percent.
  Soft budget constraints and the arrears
Soft budget constraints and weak corporate governance allowed a
faster wage growth, especially in 1995, 1996 and 1998.
This problem is acute in the case of major utilities. Despite the
losses and arrears, the wages in this sector remain some of the
highest in Romania.
   Exchange rate regime liberalization
1992: the actual exchange rate regime was introduced. Its main
   features are full retention regime and the convertibility of the Leu
Until the end of 1996 administrated controls persisted, which led to
   the existence of three exchange rates in parallel.
January 1997: a managed float rate was introduced.
Mid-1999: Romania faced some difficulties with reimbursement of
   the foreign debt. The NBR had to interfere in order to stop the
   excessive depreciation of the Leu.

  Foreign trade liberalization
The policies adopted aimed the geographical diversification and
restructuring of exported and imported commodities.
Trade liberalization is relatively high in Romania.
Empirical Analysis of the Real Exchange Rate
   Determinants: The Case of Romania
                 ln rer  f (ln tnt, ln w, nfa, open)
                                 -     - -        +
  lnrer –
      –the real wageexchange rate
 open––netThe realprice of non-traded to traded goods
  nfa   openness
            relative
 lntnt theforeign assets
  lnw
    The tradables pricesare rate between the nominal assets exports
     Net foreign assets definedapproximated by foreignthe prices,
     The openness is the ratio the defined imports wage minus
            calculated as were as as sum of as
  The isreal exchange defined is the total the non-food nominal
     It                                                      and in the
  exchange nontradables consumer price GDP.the as the weighted
    while the to and the prices were with
  national economy ROL/USD deflatedcomputed GDP.
 (expressed inrate foreigners, expressed on a ratio to consumer price
  total liabilitiesnational currency) scaled as index.
    average of food and in the index means
                                      prices.
  index. foreign assets toservices nominal wageof a real establishedthe
            The decrease measure the effects is often appreciation
     Openness is economies theas a measure of trade regime on by
                      used appear
     Net
     In transition                                     country’s external
  of ratio An with intoof and the direct toin worker’s nominal
 Thisthe Leu taken respect to mayorder toquotation. measure of
  collective bargaining. Exogenous increases the as the Balassa-
                   increase account in be interpreted
 equilibriumisexchange ratenfa certainly leads measureappreciation of
  position.                                                a
  wage determine inflation pressuresthat transition net
  the degree of effect. This means in the of More open regimes are
      national foreign the liberalization. the countries assets
 theSamuelson currency; trademaintenance economy. foreignwhere
  stock allows to more depreciatedhave a higher inflation rate. CPI
    productivity the central will currency.
 associated with rises fasterbank to support the domestic currency.
    increase has a result the real appreciation of the domestic
    currency.
The time series used
   4.608                                                    .8

   4.607                                                    .6

   4.606
                                                            .4
   4.605
                                                            .2
   4.604
                                                            .0
   4.603

   4.602                                                    -.2

   4.601                                                    -.4
           1994 1995 1996 1997 1998 1999 2000 2001                1994 1995 1996 1997 1998 1999 2000 2001

                             LNRER                                                  LNTNT



     5.0                                                     8


     4.9                                                     6


     4.8                                                     4


     4.7                                                     2


     4.6                                                     0


     4.5                                                    -2
           1994 1995 1996 1997 1998 1999 2000 2001                1994 1995 1996 1997 1998 1999 2000 2001

                              LNW                                                    NFA



                                 3.6

                                 3.2

                                 2.8

                                 2.4

                                 2.0

                                 1.6

                                 1.2

                                 0.8
                                       1994 1995 1996 1997 1998 1999 2000 2001

                                                         OPEN
                  Unit root tests
In order to test the stationarity of time series I
applied Perron’s tests (1994)


                       Order of      Level of
        Variables      Integration   Significance
          lnrer        I(1) TC         1%
          lntnt        I(1) TC         1%
          lnw          I(1) TC         1%
          nfa           I(1) T         1%
          open          I(1) T         1%
                   Lag Length Tests

Lag       LogL       LR          FPE         AIC          SC           HQ

      0   346.8935   NA          3.97E-10    -7.458281    -7.038848    -7.28922
      1   947.1214   1092.55     9.68E-16    -20.38475    -19.26626    -19.93392
      2   1008.546   104.9045    4.30E-16    -21.20328    -19.38573*   -20.47068*
      3   1034.986   42.18598    4.24E-16    -21.23565    -18.71905    -20.22128
      4   1066.187   46.27517*   3.81E-16*   -21.37499*   -18.15934    -20.07885
      5   1086.55    27.91319    4.44E-16    -21.27078    -17.35608    -19.69288
      6   1115.5     36.4315     4.38E-16    -21.35955    -16.74579    -19.49987
      7   1127.766   14.05728    6.49E-16    -21.07338    -15.76057    -18.93194
      8   1150.519   23.52048    7.91E-16    -21.0229     -15.01103    -18.59969
               Cointegration test
Hypothesized                  Trace                Critical value
No. of CE(s)   Eigenvalue     Statistic      5 Percent      1 Percent

None **            0.427573      104.9282          68.52            76.07
At most 1 *        0.252823      53.60422          47.21            54.46
At most 2          0.180289      26.79056          29.68            35.65
At most 3          0.069478      8.500592          15.41            20.04
At most 4          0.020182      1.875694           3.76             6.65

Hypothesized                  Max-Eigen            Critical value
No. of CE(s)   Eigenvalue     Statistic      5 Percent      1 Percent

None **            0.427573       51.32398         33.46            38.77
At most 1          0.252823       26.81366         27.07            32.24
At most 2          0.180289       18.28997         20.97            25.52
At most 3          0.069478       6.624898         14.07            18.63
At most 4          0.020182       1.875694           3.76            6.65
             Cointegrating relationship
     Normalization with respect to the real exchange rate yields the
     following cointegrating relationship:
ln rer  4.680653  0.014566  ln tnt  0.022180  ln w  0.001637  nfa  0.016266  open
                      (0.00362)              (0.00702)                   (0.00067)   (0.00269)


                           .015

                           .010


                           .005


                           .000

                          -.005


                          -.010

                          -.015
                                  1994 1995 1996 1997 1998 1999 2000 2001

                                                Cointegrating relation
Impulse Response Function
                            Response to Cholesky One S.D. Innovations ± 2 S.E.

                 Response of LNRER to LNRER                                          Response of LNRER to LNTNT
    .0005                                                              .0005

    .0004                                                              .0004

    .0003                                                              .0003

    .0002                                                              .0002

    .0001                                                              .0001

    .0000                                                              .0000

    -.0001                                                         -.0001

    -.0002                                                         -.0002

    -.0003                                                         -.0003
             2     4    6     8        10       12        14                    2        4        6   8   10   12   14




                  Response of LNRER to LNW                                               Response of LNRER to NFA
    .0005                                                              .0005

    .0004                                                              .0004

    .0003                                                              .0003

    .0002                                                              .0002

    .0001                                                              .0001

    .0000                                                              .0000

    -.0001                                                         -.0001

    -.0002                                                         -.0002

    -.0003                                                         -.0003
             2     4    6     8        10       12        14                    2        4        6   8   10   12   14




                                                     Response of LNRER to OPEN
                                  .0005

                                  .0004

                                  .0003

                                  .0002

                                  .0001

                                  .0000

                                  -.0001

                                  -.0002

                                  -.0003
                                            2         4        6   8       10       12       14
Variance Decomposition
                                                     Variance Decomposition

         Percent LNRER variance due to LNRER                                Percent LNRER variance due to LNTNT
   100                                                                100


   80                                                                 80


   60                                                                 60


   40                                                                 40


   20                                                                 20


    0                                                                  0
         2       4     6     8         10       12       14                 2       4     6     8    10    12       14




             Percent LNRER variance due to LNW                                  Percent LNRER variance due to NFA
   100                                                                100


   80                                                                 80


   60                                                                 60


   40                                                                 40


   20                                                                 20


    0                                                                  0
         2       4     6     8         10       12       14                 2       4     6     8    10    12       14




                                            Percent LNRER variance due to OPEN
                                 100


                                 80


                                 60


                                 40


                                 20


                                  0
                                            2        4        6   8    10   12       14
Weak Exogeneity Tests

  Variables   χ2(1)    Probability
    lnrer     0.8247     0,3697
    lntnt     0,1666     0,6831
    lnw       0,1733     0,6772
     nfa      7.9714     0,0048*
    open      3.1030     0,0755
                         Short-Run Dynamics

           Vector error correction model, after imposing
           restrictions on the net foreign assets’ coefficients:

 ln rer  0.0257  [ln rer (1)  0.014566  ln tnt (1)  0.022180  ln w(1)  0.001637  nfa (1)
           0.016266  open (1)  4.680563 ]  0.349784   ln rer (1)  0.448357   ln rer (2)
           0.24764   ln rer (3)  0.002   ln tnt (2)  0.001887  lnw(1)  0.002275  lnw(-2)
           0.001829  lnw(-3)  0.000369  open (2)  0.000364  dummy 97  0.000134  dummy 99
              CONCLUSION

The managed float exchange rate regime seems to
be the most appropriate policy for Romania.

This paper argues in favour of policies that will help
increase the flexibility of wages and reduce the
monopoly power in order to contribute to
endogenous adjustment of the real exchange rate.