Abstract：This study investigates the impact of the purchasing power parity (PPP) adjusted real exchange rate on economic growth across OECD and non-OECD countries using panel data spanning five decades. The analysis employs various econometric models, including mixed-effects, random-effects, fixed-individual-effects, and fixed-time-effects models. The findings reveal a significant positive effect of the PPP-adjusted real exchange rate on economic growth, particularly in OECD countries, where a 1% increase in the exchange rate corresponds to a 0.022% increase in economic growth. In contrast, the effect in non-OECD countries is relatively weak, with a coefficient of 0.003, indicating that these economies are more dependent on factors such as capital accumulation and investment for growth. The study also highlights the importance of control variables, including total factor productivity, human capital, capital stock, government consumption, trade openness, and foreign investment, in driving economic growth. The analysis concludes that while the real exchange rate plays a crucial role in developed countries, its significance diminishes when accounting for time effects. Furthermore, it underscores the need for tailored policy measures that consider the unique economic structures and growth drivers of OECD and non-OECD nations. The study contributes to the literature by providing insights into the long-term relationship between exchange rates and economic growth and suggests avenues for future research.
Keywords-Purchasing Power Parity；Real Exchange Rate；Economic Growth；
1. Introduction
This study explores the impact of PPP-adjusted real exchange rates on economic growth, comparing OECD and non-OECD countries. By eliminating price level differences, PPP-adjusted rates offer a more accurate reflection of economic conditions, making them critical for cross-country comparisons. The research hypothesises that OECD countries, with stable monetary policies and stronger economic structures, will exhibit higher real exchange rate stability, which supports steady economic growth. In contrast, non-OECD countries may experience greater volatility due to external shocks and policy instability, potentially hindering their growth.
The motivation for this research stems from the central role of PPP in international economics. While the concept of PPP suggests that exchange rates should adjust to equalize the cost of a basket of goods across countries, deviations from this ideal frequently occur. These deviations raise questions about their underlying causes and how they affect the economic growth of different regions. The literature has largely focused on developed countries, leaving a gap in understanding the dynamics in non-OECD countries, where inflation and commodity exports play a larger role.
This study aims to fill that gap by providing a comparative analysis of real exchange rate stability and economic growth between OECD and non-OECD nations. The results will offer insights into how structural differences between developed and developing economies influence exchange rate behavior and growth outcomes.
2.Literature review on the Purchasing Power Parity (PPP) Hypothesis
The Purchasing Power Parity (PPP) hypothesis is a foundational theory in economics concerning the equilibrium value of currencies. Introduced by Cassel (1918), the hypothesis asserts that the exchange rate between two currencies is determined by the ratio of price levels in each country. Cassel proposed that, under conditions of comprehensive free trade, actual exchange rates should not deviate significantly from their purchasing power parity. Despite early critiques by economists such as Keynes (1923) and Haberler (1945), who questioned the mechanisms of exchange rate determination and the impacts of monetary disturbances, the PPP hypothesis has persisted as a focal point of economic debate.
The theory is categorized into absolute and relative PPP. The absolute version posits that the exchange rate is defined by the ratio of domestic to foreign price levels at a specific time. Conversely, relative PPP redefines exchange rates and price levels as index numbers, which represent the ratio of current values to base-period values. This relative approach is primarily examined over the long term, encompassing periods from the gold standard of the 19th century to the managed float system of the 1970s.
Understanding nominal and real exchange rates is crucial for testing the PPP. The nominal exchange rate reflects the price of one currency in terms of another, while the real exchange rate adjusts for relative price levels. The real exchange rate is pivotal in assessing PPP as it accounts for non-stationarity. A non-stationary real exchange rate indicates a long-term deviation from equilibrium due to persistent shocks, while a stationary rate suggests that disturbances will dissipate over time, allowing a return to long-run equilibrium (Dornbusch & Krugman, 1976; Rogoff, 1996). The ongoing discourse on short- and long-term deviations of the real exchange rate remains significant in exchange rate economics.
Rogoff (1996) emphasized the considerable volatility of real exchange rates in the short run and the slow mean reversion, arguing that PPP cannot adequately explain short-term fluctuations in exchange rates (Frenkel, 1981). Some studies contend that deviations in real exchange rates are permanent due to persistent demand and supply shocks (Roll, 1979; Stockman, 1980; Adler & Lehmann, 1983), while others reject the notion of a random walk in real exchange rates over extended periods, particularly post-World War II when fluctuations were often temporary (Huizinga, 1987; Grilli & Kaminsky, 1991).
In terms of economic growth, traditional views suggest that currency depreciation positively influences growth by enhancing net exports. Depreciation lowers the relative price of domestic goods and increases the price of foreign goods, which can lead to a rise in export volume and, subsequently, economic growth. Consequently, currency devaluation is frequently employed as a policy tool to stimulate growth, with numerous studies corroborating the positive correlation between exchange rate increases and economic growth (Rodrik, 2008; Missio et al., 2015; Habib et al., 2017). Rodrik (2008) found that real exchange rate depreciation had a beneficial effect on growth in developing countries, though this relationship was not observed in advanced economies. Missio et al. (2015) demonstrated that a competitive real exchange rate significantly influenced growth in 63 developing countries from 1978 to 2007. Similarly, Habib et al. (2017) showed that real depreciation led to increased annual real GDP growth across over 150 countries following the Bretton Woods period.
Contrastingly, structuralist economists argue that rising exchange rates can hinder economic growth, especially in developing nations that heavily depend on imported inputs such as raw materials and intermediate goods. For these economies, currency depreciation may restrict imports, thereby limiting access to essential production inputs and potentially stifling growth. Additionally, imports are crucial for technology transfer to developing countries, which enhances their technological capabilities and productivity. Therefore, while devaluation can stimulate growth in certain contexts, it may obstruct economic development in countries with significant import dependencies (Bird & Rajan, 2004; Hallwood & Macdonald, 2003).
In the context of the PPP hypothesis, converting exchange rates and GDP deflators into logarithmic form allows for a more effective analysis of percentage changes, enhancing clarity regarding elasticities and mitigating data variability. This transformation also addresses non-linearity issues, facilitating a more accurate examination of deviations from PPP over time. Overall, this methodological approach contributes to a deeper understanding of the dynamics underlying exchange rates and their long-term equilibrium relationships.
Despite the extensive exploration of the PPP hypothesis, significant gaps remain in the literature. Many studies predominantly focus on developed economies, leaving a paucity of research addressing the implications of PPP in emerging and developing markets. Additionally, while the relationship between exchange rates and economic growth has been established, the mechanisms driving this relationship, particularly the role of external shocks and structural vulnerabilities in these economies, require further investigation. Furthermore, the long-term stability of the real exchange rate in the context of recent economic developments, including globalization and technological advancements, is an area ripe for exploration. Understanding these dynamics is essential for crafting effective economic policies tailored to the specific contexts of diverse economies.
3. Methodology
3.1. PPP Theory
3.1.1 Generalization of PPP theory
Theory of purchasing power parity is based on the principle of the law of one price. At the same point in time, the exchange rate should depend on the purchasing power of the currencies of the two countries. The value of the same item in different countries should be the same, and if the same currency is used, the price should be the same. When there is a mismatch in purchasing power parity, if the same currency is used, the price of a commodity in the two countries is different. Due to the increase in demand, the price of the commodity in the country with the lower price will rise, and it will eventually return to the theoretical price, once again conforming to the theory of purchasing power parity. At the same point in time, the exchange rate under the PPP theory should be the ratio of the price index of the home country to that of the foreign country. The price index reflects the overall price level of the country. As shown in Equation (1).
The real exchange rate indicates the relative price of foreign goods and domestic goods, reflecting the relative competitiveness of domestic goods. The real exchange rate formula is the product of the nominal exchange rate and the ratio of the price indicators of the two countries, as shown in Equation (2).
According to formulas (1) and (2), under the purchasing power parity theory, when the nominal exchange rate is also the ratio of the price indicators of the two countries, the real exchange rate is 1.Under this condition, the price of the same commodity purchased in different regions should be the same.
The theory of purchasing power parity (PPP) is of great importance to economics. Particularly in international trade, PPP provides a theoretical basis for measuring the purchasing power of currencies in different countries.PPP usually reflects long-term trends more accurately, and is particularly applicable to countries with lower levels of economic development and less international trade, where currency exchange rates may be more susceptible to short-term fluctuations.
3.1.2 Purchasing Power Parity (PPP) Theory Validation
3.1.2.1 Charting real exchange rate trends
The following graph plots the trend of the real exchange rate for each country based on the calculated real exchange rate, which shows that for most countries the rer
The distribution is around rer = 1, suggesting that the data for most countries fulfil the PPP assumption.
Fig.1 RER Trends for Selected Countries
3.1.2.2 t-test
The K-S test was performed on the five-year data for each country, and the data for each country conformed to a normal distribution. In order to verify whether the mean of the real exchange rate data of each country is significantly different from 1, a t-test is then performed on the five-year real exchange rate data of each country. According to the t-test results, 81.4 per cent of the countries support the absolute PPP theory. An overall t-test on the five-year average data for all countries shows (t-statistic of 1.0005, p-value of 0.3176) that the mean of the real exchange rate (RER) is not significantly different from one. Overall, the t-test results support the absolute PPP theory The blue area in the figure shows the country's construction that has passed the inspection.
Fig.2 Countries with absolute PPP theory supported by data
3.2. Data
3.2.1. Data source
In this study, we use Penn World Table version 10.01(https://www.rug.nl/ggdc/productivity/pwt) to assess the performance of our proposed model. The data provides information on relative levels of income, output, input and productivity, covering 183 countries between 1950 and 2019.
3.2.2. Data processing
Based on existing economic theories, we retain some original variables for feature engineering, which is explained in greater detail in 3.3.1 Variable selection and Definition. The following are descriptions of these variables and the features created based on them, along with the calculation formulas for the non-original variables.
Variable | Obs | Mean | Std. Dev. | Min | Max |
rgdpna | 5100 | 443411.05 | 1163915.8 | 719.208 | 20572606 |
rer | 5100 | 252778.5 | 17951636 | 0 | 1.282e+09 |
tfp | 5100 | 1.058 | .487 | .308 | 9.397 |
hc | 5100 | 2.222 | .698 | 1.007 | 4.352 |
cs | 5100 | 1763200.2 | 4651165.1 | 981.858 | 99608664 |
gc | 5100 | 24666552 | 1.733e+08 | 121.743 | 2.104e+09 |
td | 5100 | 311.74 | 1634.175 | .046 | 21825.507 |
invest | 5100 | 64166702 | 4.578e+08 | -7696321.5 | 6.254e+09 |
Growth | 5099 | 0 | .452 | -7.494 | 4.714 |
lnRER | 5099 | .003 | 1.1 | -28.625 | 6.767 |
lnTFP | 5099 | 0 | .085 | -1.056 | 2.233 |
lnHC | 5099 | 0 | .089 | -1.393 | .624 |
lnCS | 5099 | 0 | .516 | -8.166 | 5.239 |
lnGC | 5099 | -.001 | .616 | -12.688 | 9.453 |
lnTD | 5099 | -.001 | .542 | -8.553 | 8.851 |
lnInvest | 5093 | -.002 | .675 | -13.923 | 9.185 |
Table 1: Original and created features
Table 2: Formulas of created features,where Xr is nominal exchange rate, Pdomestic is Price level in the domestic economy, Pforeign is Price level in the foreign economy, Pl_con is price level of CGDPo (price level of USA GDPo in 2017=1), q_x is exports at constant national 2017 prices, q_m is imports at constant national 2017 prices.
We remove data from the dataset prior to 1970 due to excessive missing values. Based on the temporal trends of each variable across countries, there is no reasonable method available to impute such a significant number of missing values. We also remove data from countries which contain excessive missing values in the features we filtered or created. Specifically, these countries are ARM, CZE, EST, HRV, KAZ, KGZ, LTU, LVA, MDA, RUS, SVK, SVN, TJK, UKR, SRB. We use post mean interpolation to replace remaining missing values in tfp.
Compared to the original variables, the log-transformed data is more stable and exhibits no significant trends, thereby meeting the modeling requirements, as shown by visualizations.
Fig.3 Time series plot of rgdpna and log_rgpna(growth rate)
Fig.4 Time series plot of tfp and log_tfp
Fig.5 Time series plot of gc and log_gc
3.3. Research design
3.3.1. Variable selection and definition
In terms of variable selection, drawing on existing literature and considering data availability, the model was constructed using economic growth as the dependent variable, the real exchange rate as the core independent variable, and other variables such as technological progress, human capital, capital stock, government expenditures, degree of openness of the country, foreign investment, etc., as control variables (table 3). These control variables help to explain the drivers of economic growth and reduce errors and biases in the direct impact of the real exchange rate on economic growth.
The theory of purchasing power parity (PPP) states that exchange rates between different countries should reflect the price levels of the same goods and services in both countries. Rodrik (2008) states that undervalued real exchange rates can increase national exports and investments, thus driving productivity and economic growth.
Technological progress is regarded as an exogenous variable in neoclassical growth theories (e.g. the Solow-Swan model), which determines productivity gains that lead to a more efficient use of resources, improved product quality and lower production costs, further boosting economic growth (Solow, 1956).
The human capital theory, put forward by economists such as Schultz (1961) and Becker (1962), suggests that the level of education, health and accumulation of skills are the main components of human capital, and that a high level of human capital can raise labour productivity, promote innovation and technological progress, and enhance the long-term growth potential of the economy.
The amount of capital stock is one of the key input factors in neoclassical growth models and is usually represented by physical capital (e.g. equipment, infrastructure, etc.) (Solow, 1956). Increasing the capital stock, i.e., increasing investment in infrastructure development, factories, equipment, etc., can enhance a country's productive capacity, which in turn leads to incremental increases and improves the efficiency of the economy in aggregate.
According to Keynesian economics, the government can directly stimulate aggregate demand by increasing consumer spending, thereby promoting economic growth (Seidman, 2012). In times of recession or downturn, government spending can fill the gap of insufficient private sector demand, boosting employment and production.
According to the Heckscher-Ohlin model (Heckscher et al., 1991) and the theory of comparative advantage (Costinot, 2009), international trade can drive economic growth through the efficiency of resource allocation, the spread of technology and the growth of economic scale. Countries with higher openness are better able to utilise global market resources and technology, expanding export markets and gaining productivity on a larger scale. Higher openness also leads to substitution effects of technology, which promotes the upgrading of domestic industrial structure and focuses on driving economic growth.
Modernisation theory suggests that economic development is a gradual evolutionary process (Gwynne, 2009). Within the framework of modernisation theory, FDI can enhance a country's economy through capital inflows, technology transfer, sharing of management experience and market competition.
Table 3: Sample variables and description
3.3.2. Research programme
This paper adopts a quantitative analysis method and selects panel data from 1970-2019 containing OECD countries and non-OECD countries in order to explore the impact of PPP-adjusted exchange rate on economic growth. The empirical study consists of three main parts: first, the data are subjected to unit root tests and stepwise regression analyses of the variables to ensure the validity of the model; second, for the panel data, regressions are carried out using the mixed-effects model, the random-effects model, the fixed-effects model, and the double-fixed-effects model, respectively, to select the optimal model; and finally, the optimal model is used to estimate the PPP-adjusted exchange rate effect on economic growth in OECD countries and non-OECD countries respectively for comparison. The basic models used in the study are as follows:
where i denotes a different country and t denotes a year， denotes unobservable individual effects and represents the error term.
4. Empirical process and results
4.1. Model testing
4.1.1. Fisher Test
Since the panel data contains time series data, it is necessary for the sample data to be smooth in the econometric analyses to ensure the validity of the model and to avoid pseudo-regression, hence the unit root test is performed, and Fisher's test is being applied to the panel data.
Based on the results of Fisher's test it can be seen that the p-value of all the variables except is 0, which rejects the original hypothesis and is smooth, whereas the p-value of is 0.8249, which is non-smooth. represents the first-order differencing of , which has a p-value of 0, i.e., it is smooth. However, when is regressed against the results are significant and is not significant when regressed against . This may be due to differencing thereby causing the data to lose long term information. Therefore, further cointegration tests are to be conducted.
Table 4: Fisher test results
4.1.2. Cointegration test
Since the main study is the impact of exchange rate on economic growth after PPP adjustment and the variable representing human capital index is not smooth, cointegration test is conducted for these three variables.
According to the Pedroni test, in the results of the panel cointegration regression model with individual fixed effects and time trend, only individual fixed effects and neither, the p-value is 0, which rejects the original hypothesis and indicates that there is a cointegration relationship between the variables, i.e., it indicates that the human capital index maintains a long-run equilibrium relationship with the real exchange rate and economic growth. Considering that this paper studies the long-term impact of the exchange rate on economic growth after PPP adjustment, this variable can still be used as a control variable although it is non-stationary.
Table 5: Cointegration test results
4.1.3. Stepwise regression analysis
After the correlation analysis of all the variables, except for the dependent variable and the main independent variable, the other variables were ranked in order of the magnitude of the correlation coefficients with the dependent variable and stepwise regression analyses were carried out in this order to verify the validity of the control variables.
According to the results of correlation analysis, the order of correlation between the control variables and Growth is lnCS, lnHC, lnGC, lnInvest, lnTD, lnTFP. Firstly, the regression is carried out by using , which represents economic growth, as the dependent variable, and and as the independent variables, and then the other variables are added step by step in the order of correlation to obtain Models 1-6. However, the addition of the the variable of the country's openness level resulted in the human capital index being insignificant, while the subsequent addition of the variable of technological progress resulted in both being significant again. Therefore, the additional elimination of and separately, as well as the elimination of both, resulted in Models 7-9. Although the models with the elimination of both variables were also significant, the best fit of all the models was Model 6, which was the model with all variables included. Therefore, the model used in the following study is still the original model.
Table 6: Correlation
Table 7: Stepwise regression results
4.2. Model regression results
4.2.1. Comparing regression results from different regression models
The panel data were regressed using the mixed-effects model, random-effects model, fixed-individual-effects model, and fixed-time-effects model, respectively, to compare the results of the regression analyses of each model.
The regression coefficient of the real exchange rate on economic growth under the mixed effects model (OLS) and random effects model (RE) is 0.0054 and the significance level is very high (p-value close to 0). This indicates that there is a positive and significant effect of the PPP-adjusted real exchange rate on economic growth, and the effect is relatively small, with each 1% increase in the exchange rate increasing economic growth by about 0.0054%. In the fixed effects model FE1, controlling for country effects, the results are consistent with OLS and RE, and the coefficient of lnRER remains 0.0054 and significant. However, in the fixed effects model FE2 (controlling for country and year effects), the coefficient on the real exchange rate is 0.0012 but is no longer significant (p-value of 0.4131), suggesting that after controlling for the time effect, the real exchange rate is no longer significant in explaining economic growth.
This may imply that the time effect plays a larger role in explaining the relationship of real exchange rate on economic growth. The time effect usually reflects some common trends in the economy throughout the time period under study, such as the global economic cycle, changes in monetary policy, and so on. The real exchange rate may play a minor role in the context of these macro factors, and thus it loses significance in models with fixed time effects. This may imply that the global economic environment plays a greater role in the country's economic growth over certain time periods, overriding the direct impact of the exchange rate on the economy. In addition, it may also mean that the short-term impact of the PPP-adjusted real exchange rate on economic growth may be more limited after accounting for country and time heterogeneity.
Table 8: Regression results from different regression models
4.2.2. LM test and Hausmann test
adj. R-squared shows that the goodness of fit of all models is very high, ranging from about 0.9648 to 0.9656, which suggests that the models are able to explain the variations in economic growth relatively well. Since only the goodness of fit is compared, the mixed effects model has the highest goodness of fit as well as the fixed time effects model, which is not much different from the fixed individual effects model, and there is no R-squared for the random effects model, the LM test and Hausman test are additionally performed.
The LM test and Hausmann test in the experimental results are not significant, indicating that the mixed effects model is better than the random effects model, which in turn is better than the fixed effects model. This implies that the mixed effects model can better explain the individual differences and random effects in the data. In panel data analyses across countries or regions, the mixed effects model better captures the heterogeneity between countries. This suggests that the real exchange rates of different countries may have different mechanisms at work, and a unified fixed-effects model may ignore these heterogeneities, while a mixed-effects model can cope with this complexity better.
LM test: | Hausman test: |
chibar2(01) = 0.00 | chi2(7) =8.13 |
Prob>chibar2 = 1.0000 | Prob>chi2 = 0.3214 |
Table 9: LM test and Hausman test
Considering all the circumstances, fixed time effects lead to insignificant real exchange rate on economic growth, LM test and Hausmann test results are insignificant, in addition to the mixed effects model has the highest goodness of fit and it is inferred that the mixed effects model is the best model.
4.2.3. Model regression results conclusion
Overall, the impact of the PPP-adjusted real exchange rate on economic growth is significant, but this effect may weaken or cease to be significant after controlling for time effects. Other important control variables (e.g., technological progress, human capital, capital stock, etc.) have a strong positive effect on economic growth, with technological progress and capital accumulation being particularly important for economic growth.
4.3. Compare OECD and non-OECD countries
4.3.1. Analysis of main results
In OECD countries, the regression coefficient of PPP-adjusted real exchange rate is 0.022 and the p-value is highly significant (less than 0.001). This indicates that the PPP-adjusted real exchange rate has a significant positive impact on economic growth in OECD countries. For every 1 per cent increase in the real exchange rate, economic growth will increase by 0.022 per cent, indicating that the economies of developed countries are more sensitive to changes in the real exchange rate, and also reflecting the high degree of linkage between the economies of developed countries and the international market, where exchange rate changes can have a direct impact on economic growth through trade, investment and other channels.
In non-OECD countries, the impact of the PPP-adjusted real exchange rate on economic growth is relatively small, with a regression coefficient of only 0.003 and a p-value of 0.050, which is close to but not reaching the significance level. This suggests that the direct impact of real exchange rate movements on economic growth is relatively weak in non-OECD countries, probably because the economic structure of these countries is more dependent on other factors (such as capital accumulation and investment) to drive economic growth, and exchange rate fluctuations play a limited role in these countries.
Table 10: Regression results for OECD and non-OECD countries
4.3.2. Effects of other control variables
Total factor productivity (lnTFP) is important for economic growth in both groups of countries, but its influence is greater in OECD countries. This means that economic growth in developed countries is more dependent on technological progress and productivity gains.
Human capital (lnHC) is a strong driver of economic growth in OECD countries, while its role is relatively weak in non-OECD countries. This reflects the key role of high skills and education in driving economic development in developed countries.
-Capital stock (lnCS) has a significant positive impact on economic growth in both types of countries, and the impact is slightly larger in non-OECD countries. This implies that capital accumulation (e.g., infrastructure, production equipment, etc.) is crucial for economic growth in both developed and developing countries. In non-OECD countries, the role of capital stock is even more significant, suggesting that these countries may still be at the stage of capital-intensive economic growth.
Government consumption (lnGC) contributes more to economic growth in OECD countries than in non-OECD countries, suggesting that public spending is more efficient in developed countries.
Trade openness (lnTD) has a negative impact on economic growth in both groups of countries, especially in OECD countries, which may be related to trade imbalances or external economic shocks.
Foreign investment (lnvest) has a positive effect on economic growth in both groups of countries, with a slightly stronger effect in non-OECD countries. This demonstrates the critical position of investment in developing countries, particularly in the promotion of infrastructure and industrial development.
4.3.3. Causal analysis
OECD countries tend to have more openness to the outside world, higher trade dependence and more complex financial systems, and therefore their economic growth is more sensitive to changes in the real exchange rate. Changes in real exchange rates can have a direct impact on their exports and imports, capital flows and global market competitiveness, thereby affecting economic growth.
Most of the non-OECD countries are developing countries with relatively simple economic structures, dominant domestic markets and, in many cases, relatively strict exchange-rate management, where exchange-rate movements may be controlled or have little impact, so that the direct impact of the real exchange rate on economic growth is weak. Economic growth in these countries is more dependent on domestic factors such as infrastructure development, capital accumulation and government investment.
4.3.4. Conclusion
In OECD countries, the PPP-adjusted real exchange rate has a significant positive impact on economic growth, indicating that economic growth in developed countries is highly dependent on exchange rate changes. In non-OECD countries, on the other hand, the exchange rate does not have a significant impact on economic growth, indicating that economic growth in these countries is mainly dependent on factors such as domestic investment and capital accumulation.
5. Discussion
A panel regression analysis of the data shows that the purchasing power parity (PPP) adjusted real exchange rate has a significant impact on economic growth without controlling for time effects. However, after controlling for time effects, the positive correlation between technological progress and capital accumulation on economic growth becomes more significant, while the impact of the real exchange rate is significantly weaker.
The novelty of this study lies in the use of panel data spanning five decades, covering multiple business cycles as well as changes in global monetary policy, providing a more comprehensive understanding of the relationship between real exchange rates and economic growth. Unlike most studies that focus only on relative PPP-adjusted exchange rates, for example, this study examines the impact of absolute PPP-adjusted exchange rates on the economy. Government consumption expenditure is introduced as a key variable in the model, revealing the long-run role of fiscal policy in adjusting for exchange rate deviations, especially in non-OECD countries where government expenditure is particularly important in driving economic growth. This provides a new perspective for understanding exchange rate effects in different economies.
As with Amalia’s(2010) findings, the real exchange rate has a significant positive impact on economic growth under both mixed and random effects models, but after controlling for time effects, this impact weakens or even ceases to be significant, suggesting that global economic factors may dominate changes in economic growth to a greater extent than short-term fluctuations in the exchange rate. This suggests that the direct role of the exchange rate is relatively limited in certain economic cycles.
Despite the fact that this study uses panel data with long time horizons and multiple econometric models, there are some obvious limitations.
Data quality may vary across countries over time. In developing countries, data collection standards and accuracy may be inconsistent, which has an impact on the robustness of the results.
Although this study uses mixed-effects, random-effects and fixed-effects models to analyse the relationship between the exchange rate and economic growth, these models assume more stable external economic conditions and may fail to adequately take into account factors such as international capital flows, external economic shocks and changes in global monetary policy. In the context of high global economic uncertainty, the models may not accurately capture the impact of exchange rate fluctuations on economic growth in the short run.
This study finds that economic growth in OECD countries is significantly dependent on exchange rate volatility, while the dependence is weaker in non-OECD countries. This may not be entirely true for some of the OECD countries, especially those with strict exchange rate controls.
Future research should explore in greater depth the impact of global economic cycles and external shocks on exchange rate volatility, especially the direct effect of exchange rate changes on economic growth in the short run. In addition, the study could introduce more dynamic variables, and further improve the explanatory power of the model.
6.Conclusion
This study draws two key conclusions: (1) it mostly aligns with previous literature, showing that in OECD countries, stable real exchange rates support economic growth, while in non-OECD countries, the impact is more complex; (2) the effect of exchange rates on economic growth is not uniform and depends on factors such as technological progress and government spending, requiring tailored policy measures for different countries.
7.Future Work
Future research could explore three areas: (1) expanding the scope to include other developing regions; (2) incorporating external shocks, such as global crises or commodity price fluctuations, to analyze their impact on exchange rate volatility; and (3) further decomposing the effects of trade openness, such as export diversification and reliance on trading partners, to better understand its influence on growth in both OECD and non-OECD countries.
References
[1] Adler, M., & Lehmann, B. (1983). Deviations from Purchasing Power Parity in the Long Run. Journal of Finance, 38(5), 1471-1487.
[2] Becker, G.S. (1962) ‘Investment in human capital: A theoretical analysis’, Journal of Political Economy, 70(5, Part 2), pp. 9–49. doi:10.1086/258724.
[3] Cassel, G. (1918). The Theory of Purchasing Power Parity.
[4] Costinot, A. (2009) ‘An elementary theory of comparative advantage’, Econometrica, 77(4), pp. 1165–1192. doi:10.3982/ecta7636.
[5] Dornbusch, R., & Krugman, P. (1976). Flexible Exchange Rates in the Short Run.
[6] Frenkel, J. A. (1981). The Collapse of Purchasing Power Parities during the 1970s.
[7] Gwynne, R.N. (2009) ‘Modernization theory’, International Encyclopedia of Human Geography, pp. 164–168. doi:10.1016/b978-008044910-4.00108-5.
[8] Haberler, G. (1945). Prosperity and Depression.
[9] Heckscher, E.F. et al. (1991) Heckscher-Ohlin Trade theory. Cambridge, Mass: MIT Press.
[10] Huizinga, J. (1987). An Empirical Investigation of the Long-Run Behavior of Real Exchange Rates.
[11] Keynes, J. M. (1923). A Tract on Monetary Reform.
[12] Rodrik, D. (2008) ‘The real exchange rate and economic growth’, Brookings Papers on Economic Activity, 2008(2), pp. 365–412. doi:10.1353/eca.0.0020.
[13] Rogoff, K. (1996). The Purchasing Power Parity Puzzle. Journal of Economic Literature, 34(2), 647-668.
[14] Schultz, T.W. (1961) ‘Investment in Human Capital’, The American Economic Review, Vol. 51, pp. 1–17.
[15] Seidman, L.S. (2012) ‘Keynesian fiscal stimulus: What have we learned from the great recession?’, Business Economics, 47(4), pp. 273–284. doi:10.1057/be.2012.25.
[16] Solow, R.M. (1956) ‘A contribution to the theory of economic growth’, The Quarterly Journal of Economics, 70(1), p. 65. doi:10.2307/1884513.
[17] Stockman, A. C. (1980). A Theory of Exchange Rate Determination.
Comments