Balassa-Samuelson

Effect
ANA MILOVANOVI
NINA MILOVANOVI
The Purchasing Power Parity
Definition: At every moment the real parity between two countries is represented by the
quotient between the purchasing power of money in one country and the other. I propose to call
this parity the purchasing power parity.
Gustav Cassel
The Purchasing-Power Parity Doctrine:
A Reappraisal
Absolute interpretation of the doctrine:
PPP calculated as a ratio of consumer goods prices tends to approximate the equilibrium
exchange rates.

Relative interpretation of the doctrine:

Changes in relative prices indicate the necessary adjustments in exchange rates.
Equilibrium of Prices, Wages, and
Exchange Rates
Samuelsons Statement: the theory of competitive advantage does not guarantee a country against
balance-of-payments difficulties
Because:
Comp Adv Theory -> We would never be exporting a good i while exporting a good j if our competitive
advantage is in good j rather than in good i

However, this does not say that our current balance will balance our, or that our total BoP is at equilibrium.
Money wages abroad, W and w, together with the FX rate, R, determine completely the pattern of prices and
of productions.
The price of a good at any place equals the lowest cost of production anywhere translated into
commensurate currency units (transport and tariffs not included).
Effect: Definition
Countries with high productivity growth also experience high wage growth, which leads to
higher real exchange rates.
The Balassa-Samuelson effect proposes that a rise in wages in the tradable goods sector of an
emerging economy will result in a rise in wages in the non-tradable (service) sector of the
economy.
The accompanying increase in inflation makes inflation rates higher in more rapidly
growing economies, than in slow growing, established economies.

- Bela Balassa and Paul Samuelson (1963)

The Purchasing-Power Parity Doctrine:
Problem Description
The argument of Houthakker (1962.) about overvaluation of the U.S. dollar compared to the
German mark raises the question of what meaning can be attached to an international
comparison of exchange rates and purchasing-power parities.
Samuelson continues the problem with:
1) American and European cost-of-living were sometimes computed with different good
weightings; such index numbers should not be used together in:

2) Heavy transport costs and impediments DO exist, making price ratios not uniform,
GEOGRAPHICALLY.
The Purchasing-Power Parity Doctrine:
Assumptions
1. two-country model with international trade
2. non-traded good (services)
3. one limiting factor: labor
4. constant marginal rates of transformation
5. one country has an advantage in the production of traded goods
The Purchasing-Power Parity Doctrine:
Model
By assuming that international productivity differences are greater in the production of traded
goods than in the production of non traded goods, the currency of the country with higher
productivity levels will appear to be overvalued in terms of purchasing-power parity.
If per capita incomes are taken as representative of levels of productivity, the ratio will take a
following form:

The ratio is an increasing function of income levels.

The Purchasing-Power Parity Doctrine:
Introduction to Empirical Testing
Index number problem: the results will depend on the choice of weights - the final bill of goods
consumed in individual countries.
Gilbert and Kravis provide evidence of the relatively high cost of services in countries with higher
income levels.
The international productivity differences in the service sector are considerably smaller than in
the production of the traded goods, raising thereby the cost of services in high-income
countries.
The Purchasing-Power Parity Doctrine:
Empirical Testing
Hypothesis: higher level of service prices at higher income levels lead to systematic differences
between PPP and equilibrium exchange rates.
The Purchasing-Power Parity Doctrine:
Empirical Testing
Assumption: small rise in productivity in the service sector and uniform increase in productivity
in the traded goods sector will rise the relative price of the non-traded goods.
Productivity increases in the service sector appears to be generally smaller than the rise of
productivity in agriculture and manufacturing.
The Purchasing-Power Parity Doctrine:
Empirical Testing
PPP doctrine will hold if productivity increases and wage adjustments are identical in every
country (parallel changes in the general price level).
There will be no need for adjusting the rates of exchange.
Technological improvements and wage adjustments dont follow the same course in every
country.
Correspondingly, an intercountry comparison of changes in the general price level cannot be
used to indicate the need for modifications in exchange rate parties.
The Purchasing-Power Parity Doctrine:
Empirical Testing
Hypothesis: productivity improvements in the sectors producing traded goods are positively
correlated with the ratio of the general price index to the index of the prices of traded goods.
Results
In the presence of changes in productivity and prices in the sectors of traded and non-traded
goods, the reliance on the general price indexes for deciding on exchange-rate adjustments
appears to be misplaced.
Price indexes heavily weighted with internationally traded goods will not appropriately indicate
the need for modifications in exchange rates.
Conclusion
The emphasis on the need for the familiar models of international trade adjusted with
consideration of non-traded goods.
There is a relationship between PPP and exchange rates which can be used in judging the
overvaluation or undervaluation of a currency.
PPP and exchange rate relationship can be used to provide guidance for the international
comparison of national incomes and living standards.
The use of the exchange rates as conversion ratios will overstate the GNP of high-income
countries and understate that of low-income countries, with the degree of overstatement
increasing as income levels rise.
Balassa-Samuelson Effect in Transition
Economies: The Case of Slovenia
It is argued that on average one percent increase in labor productivity in industry and services
appreciated external real exchange rate by almost 1.5 percent in the period from 1993:1 to
2001:2.
At the same time, one percent increase in productivity differential caused about 1.7 percent
increase in CPI.
Real exchange rate is defined in external terms as the nominal exchange rate adjusted for the
price level differences between countries.
Balassa-Samuelson Effect in Transition
Economies: Transition Economies
Productivity gap is larger for tradable goods than for non-tradable goods.
Price of non-tradeable goods will be lower in less-developed countries than in industrial
countries.
The general price level will be lower in less-developed countries.
Transition Economies: Model
Transition Economies: Model
Transition Economies: Model
Transition Economies: Data
19 transition economies to construct price index: Armenia, Azerbaijan, Belarus, Bulgaria, Croatia,
Czech Republic, Estonia, Hungary, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Poland, Romania,
Russia, Slovak Republic, Slovenia, Ukraine, and Uzbekistan.
The criterion for the period of observation was the year after which the relative price of tradable
goods in terms of non-tradable goods started to consistently decline except in case of Belarus,
Romania and Russia.
The longest series runs from 1990 to 1998, while the shortest covers the period from 1995 to
1998.
The whole sample includes 122 observations.
Transition Economies: Results
Structural reforms in transition economies contributed to the real appreciation trend.
Productivity differential used to measure the Balassa-Samuelson effect had a pronounced effect
on appreciation of the real exchange rate in transition economies.
If the productivity in the two sectors within the country grows at different rates, there will be
offsetting movements in the relative price of tradable goods in terms of non-tradable goods.
The Case of Slovenia
Stacked columns represent the level of real exchange rate in each year.
The portions of columns correspond to actual contributions that each variable had to the level
of the real exchange rate in each year of the transition process.
The sum of all portions of a column and country-specific constants add up to the fitted value for
the real exchange rate level in the respective year.
The Case of Slovenia
The Case of Slovenia
As the effect of demand variables represented by the share of non-tradable consumption in
total private consumption and government consumption has remained relatively unaltered.
The main determinants of the level of real exchange rate in Slovenia were structural changes
and productivity differential.
The Case of Slovenia
The structural changes caused the increase of labor productivity in both sectors before 1996.
From 1996 onwards, the productivity differential between labor productivity in industry and
services has started to effect the real exchange rate substantially.
Thereafter labor productivity in industry has been increasing faster than in services.
For that reason, the productivity differential believed to cause the real exchange rate
appreciation via Balassa-Samuelson effect has started to affect the level of real exchange rate
relatively more than the extent of structural reforms which has diminished in recent years.
The Case of Slovenia
Slovenia: Conclusion
The existence of the Balassa-Samuelson effect has been confirmed.
Its size varies around 1.5% per year based on 1% increase of the productivity differential
between labor productivity in industry and services.
The Balassa-Samuelson Effect in Romania
This paper aims to provide estimates of the Balassa-Samuelson effect in Romania the extent to
which differences in productivity growth between tradables and non-tradables industries
explain the observed differences in inflation between Romania and the Euro Area.
Considers productivity differential of each sector and the weight of the tradables sector in
Romania in relation to the Euro Area.
Romania: Model
The international Balassa-Samuelson explains the extent to which the inflation differential
between the two countries is explained by the productivity differential between traded and non-
traded industries.
Romania: Data
The source of data is Eurostat and The National Bank of Romania databases.
The economies and periods covered are Romania (2002:Q1-2006Q4) and Euro Area (2002:Q1-
2006:Q4).
Description of variables:
Quarterly observations of value added from the production side GDP estimates(decomposed
into tradables and non-tradables) CPI rates of inflation.
Nominal exchange rates of domestic currency against the euro (quarterly averages).
Employment (quarterly averages) in tradable and non-tradable industries.
Romania: Empirical Evidence
The Balassa-Samuelson model assumes that productivity is higher in the tradable sector.
As a consequence, wages in this sector have the tendency to rise but, since labor is mobile, the
increase in salaries spreads across the whole economy
Romania: Empirical Evidence
Romania: Empirical Evidence
The wage growth was similar across the two sectors which is consistent with the theory.
Labor mobility determined the wages in the non-tradable sector to rise as much as wages in the
tradable sector, although productivity gains are smaller.
Romania: Empirical Evidence
Romania: Results
Balassa-Samuelson effect in Romania explains on average only 0.569 percentage points of the
observed inflation differential -8.06 percentage points (2002Q1:2006Q4).
Romania: Conclusion
The estimates are lower that those obtained in previous studies because we considered the
effect of the productivity differential on the inflation differential and not on domestic inflation.
The Balassa-Samuelson effect does not explain much of the observed inflation differential.
Thus factors, other than productivity differences, are responsible for the observed differences in
inflation rates.
The Case of Croatia
This paper wishes to analyze the importance of the B-S effect for Croatia, and other countries of
Central and Eastern Europe, which are new, or are hoping to become new, EU member states.

- using both monthly and annual data

- investigate the basic assumptions of the B-s hypothesis
- place a figure to the size of the long-run inflation, the inflation differential, and the real
appreciation which was driven by the B-S effect
Croatia: Theoretical Background (Model)
It is a well-understood fact that purchasing power parity (PPP) in its absolute version does not hold true for transitional
and developing countries because these countries currencies are undervalued in terms of PPP

Real Exchange Rate:

A real exchange rate higher than 1 = undervalued.

The reason for this undervaluation in terms of PPP can be usually traced back to the traditional
Balassa-Samuelson argument: the less developed country is usually less productive in producing
tradable goods.
 The model continues
This relationship can be worked out in a formal way by using a two-sector neo-classical framework with perfect capital mobility and with the interest rate assumed
exogenous.

where circumflexes (^) stand for growth rates and small letters indicate variables taken in natural logarithms. and denote the share of labour in the open and
closed sectors

-> represents the growth rate of the relative price of non-tradable goods
-> is the sectoral difference of growth rates of total factor productivity.

Change the initial equation on the basis of avg labour productivity we get:

where Y and L denote output and labour and Y L is average labour productivity (ALP).

Transform it using logarithms:

where const is a constant term containing log( ) and log( ).

Sector Breakdown
Open Sector Breakdown:

(1) manufacturing

(2) industry

(3) industry and agriculture

(4) industry, transport and telecommunications, and hotels and restaurants and finally
(5) agriculture, industry, transport and telecommunications, and hotels and restaurants.

Alternative Measures for Closed Sector Breakdown:

(1) the remaining market-based sectors

(2) the remaining market-based sectors plus real estates, (1) and (2) augmented with agriculture if not used in the open sector

(3) market-based sectors and non-market based sectors (education, health, public administration and other communal services)
(4) a measure of (3) completed with agriculture.

This yields a total of 18 combinations between open and closed sectors.

 Croatia: Estimation Techniques
Dynamic Ordinary Least Squares (DOLS):

where k1 and k2 denote, respectively, leads and lags. Max lag length set to 6.

Critical Values defined by MacKinnon

The bounds testing approach uses the error correction form of the ARDL model
Croatia: Basic Assumptions
1. Real wages are linked to productivity in the open sector
2. Nominal wages tend to equalize across sectors
3. Dual productivity is linked to the relative price of market-based non-tradable goods
4. PPP holds for the open sector
Croatia: Empirical Evidence (Yearly)
Croatia: Wages rose more slowly than productivity from 2000 to 2002, otherwise, generally
speaking, productivity and real wages broadly grew hand in hand.
Croatia: Empirical Evidence (Yearly)
Croatia - Wage Equalization: Pretty stable, except for jump-like changes can be observed, unlike
in Romania where the ratio decreased implying that nominal wages grew faster in the closed
sector than in the open sector (amplification of the B-S effect).
Croatia: Empirical Evidence (Monthly)
More rigorous assumptions
1. Productivity in the open sector is cointegrated with real wages in the open sector, with the
estimated long-term coefficient being equal to 1.
2. The sectoral wage ratio is difference stationary.
3. Dual productivity is cointegrated with the relative price of market-based nontradable goods,
with the estimated long-term coefficient being equal to 1.
4. The tradable price-based real exchange rate is difference stationary.
Croatia: Empirical Evidence (Monthly)
Passed: 1. Productivity in the open sector is cointegrated
with real wages in the open sector, with the estimated long-
term coefficient being equal to 1

Results for Croatia indicate a cointegration which can be detected

unambiguously only when the bounds testing approach (ARDL) is used. The
estimated long-run coefficient is slightly higher than 1 and increases
somewhat for the period from 1998 to 2004.
Croatia: Empirical Evidence (Monthly)
Failed: 2. The sectoral wage ratio is difference
stationary.
The sectoral wage ratio is defined as the ratio of nominal
gross wages in industry to those in the whole economy.
The Augmented Dickey Fuller (ADF), the Phillips-Perron (PP)
and the Elliott-Rothenberg-Stock (ERS) point optimal unit
root tests are unable to reject the presence of a unit root,
while the Kwiatowski-Phillips-Schmidt-Shin (KPSS) test
mostly rejects stationarity for the whole sample and for a
shorter period, i.e. 1996 to 2004.
Note also that the wage ratios based on both gross and net
monthly wages exhibit trend stationarity for the subperiod.
In sum, all series, including Croatia, either have a unit root
or are trend stationary, implying the first and/or second
moments to be unstable over time.
Croatia: Empirical Evidence
(Monthly)
Semi-Passed: 3. Dual productivity is cointegrated
with the relative price of market-based nontradable
goods, with the estimated long-term coefficient
being equal to 1.
In the results, we can observe that productivity and relative
prices based on service prices (SERVGOODS or SERVPPI).
CPI-to-PPI ratio can be viewed as (services and goods)/goods
while the two other variables are constructed as services/goods
Croatia: No cointegration is found for Croatia. The CPI-to-PPI
ratio (CPIPPI) rescues Croatia as there seems to be a positive
relationship for Croatia (as opposed to the no-cointegration
finding for SERVGOODS and SERVPPI). . Finally, the OLS estimates
of the first different data are systematically insignificant or have
the wrong sign.
Croatia: Empirical Evidence (Monthly)
Failed: 4. The tradable price-based real exchange rate is difference stationary.

Croatia: From the results, it can be seen that the PPI-based real exchange rate is clearly not difference stationary in
levels.
 Croatia: Conclusion?
They concludes that the results indicate that the B-S effect could POSSIBLY work well in Croatia

However, what is the influence of the B-S effect on overall inflation?

Also, the results indicate that relative PPP is rejected for the real exchange rate of the open
sector, which implies that the B-S effect will not be able to explain the entirety of real exchange
rate movements.
* The B-S effect could then provide an explanation for some of the changes in the difference between the (CPI-based) overall real exchange rate and the real
exchange rate of the open sector.
Croatia: B-S effect on Inflation Rates

where (1 ) is the share of non-tradables in the consumer basket

1 conceptually corresponds to the estimated coefficient from Tables 4a and 4b, which connects the relative
price of non-tradables to productivity, and which, ideally, should be 1.
PROD is the average labour productivity in the tradable (T) and non-tradable (NT) sectors.
 Croatia: B-S effect on Inflation Rates
Table 6a
Croatia appears to be less influenced
by the choice of sectoral classification.
Table 8
Results indicate that the B-S effect may be negative for
Croatia in the period of 1992 2002 when using data based
on national accounts.
However, industrial production-based figures indicate a
positive effect. This is mainly because such figures do not
take account of productivity increases in services.
Nevertheless, the effect rises to about 0.8 percentage point
in Croatia for the period of 19962002.
Croatia: Equilibrium Real Appreciation
When adjusting the figures in Table 8 with a foreign benchmark (avg of prior 3 studies: 1.2% for
19922003 and 1.0% for 19962003) we find that for Croatia:

the direction of a change in the equilibrium exchange rate hinges on whether or not national
accounts or industrial production-based data.

However, using data obtained from national accounts seems more appropriate for measuring
the B-S effect:

This would imply an equilibrium appreciation in Croatia.

Croatia: Final Conclusion
Analysis of the basic assumptions of the B-S effect reveals:
that the pass-through from productivity gains in the open sector to the relative price of non-tradable goods is not
proportionate
Because:
(a) real wages are not proportionately linked to productivity in the open sector
(b) the wage equalization process across sectors is far from perfect
(c) the relative price of non-tradables rises quicker or slower than productivity gains even when taking account for the
imperfect functioning of the two other assumptions.

-> The Balassa-Samuelson effect is found to play only a slightly than limited role for overall inflation and real exchange
rate determination.
Other external factors are deemed more important than the B-S effect.
 Comparison to Developed G7 Countries
The results of the empirical application of the Samuelson-Balassa hypothesis for G-7 countries
suggest that relative prices of non-tradable to tradable goods are explained, in part, by
productivity. Increases in tradable sector relative to non-tradable sector productivity increase the
relative prices of non-tradables.
Faster growth in productivity in the traded sector, relative to productivity growth in the
nontraded sector, increases the relative price in the nontraded good.
References
1. Balassa, B. (1964) The Purchasing-Power Parity Doctrine: A Reappraisal
2. Samuleson, P. (1964) Theoretical Notes on Trade Problems
3. Dedu, V. ; Dumitrescu B. (2010) The Balassa-Samuelson Effect in Romania
4. Jazbec, B. (2002) Balassa-Samuelson Effect in Transition Economies: The Case of Slovenia
5. Egert, B. (2005) Balassa-Samuelson Meets South Eastern Europe, the CIS, and Turkey: The
Close Encounter of the Third Kind?
6. Coto-Martinez, J. ; Reboredo, J. (2003) The Balassa-Samuelson Effect in an Imperfectly
Competitive Economy: Empirical Evidence for G7 Countries
Thank you for your
attention!
Back up slides
Introduction to Empirical Testing
Romania: Model
Aggregate price level is decomposed into a traded and non-traded component both domestically
and abroad:

The real exchange rate is defined:

Then based on these relations it is obtained:

Romania: Model
Law of one price holds in the tradable sector:

Next we determine the relation between the change in relative prices and the productivity
differential between traded and non-traded sector
Small open economy and a Cobb-Douglas production function:
Romania: Model
The profit functions for both economies:

Necessary and sufficient conditions for profit maximization:

Romania: Model
Thus, we have:

The relative price of non-tradable versus tradable will rise if the productivity in the tradable
sector is higher.
Romania: Empirical Evidence
Romania: Empirical Evidence