Printer Friendly

Common stocks and inflation: an empirical analysis of G7 and BRICS.

Introduction

When the Fisher effect is applied to the stock returns/inflation relationship, it is suggested that stocks would move one-to-one with inflation, thus making stocks a hedge for inflation. Results of past research on the relationship between stock returns and inflation were mixed. Ongoing studies examine whether the two move one-to-one or imply that stocks are a hedge for inflation, or if the relationship is an inverse one, meaning stocks would not provide any protection against inflation. Overall, a section of the literature is concerned with the relationship between stock returns and inflation in different countries with different monetary and inflationary regimes, and other sections of literature are dedicated to testing theories put forth to explain the nature of the relationships.

In an attempt to understand this relationship and extend existing literature, this study proposes testing the Fisher effect in two groups of countries, the G7 group of low inflation industrialized countries (Canada, France, Germany, Italy, Japan, United Kingdom, and United States) and a group of relatively high inflation emerging countries including Brazil, Russia, India, China and South Africa (BRICS). By using two different groups of countries, the study examines if a one-to-one movement of the variables relates to the economy's inflation rate in the short-run and consumer price index in the long-run.

It is expected that if a country's inflation rate influences the short-run relationship between inflation and stock returns, then the results from the G7 group of low inflation countries should be different from the BRICS group of relatively high inflation countries. This article examines the short-run relationship between stock returns and inflation and uses cointegration and vector error correction models to study the long-run relationship between the consumer price index and stock prices.

Literature Review

Important to the stock return/inflation relationship is the Fisher effect (Fisher 1930), which, when generalized to this topic, suggests that stocks returns and inflation will move one-to-one. Some researchers have found a negative relationship between the two suggesting the theory does not hold in this context. Many economists have set forth theories as to why this negative relationship occurs.

The proxy hypothesis states that a negative relationship between stock returns and expected inflation is observed because stock prices and expected inflation react in opposite ways to changes in expected real economic activity. For example if real economic activity is expected to be high, stock prices rise, but the demand for money also goes up. If it is not matched by an equal increase in money supply, then inflation will decrease (see Fama 1981, Gallagher and Taylor 2002).

The demand for money is not the only factor inducing a negative relationship between stock returns and inflation; changes in the money supply can also produce a similar effect. That is, a fall in stock prices signals a drop in economic activity and government tax revenue. Given fixed expenditures, this leads to an expectation that the government will run a budget deficit, and may have to take inflationary measures to finance the deficit. In summary, a decrease in stock prices will be associated with an increase in Treasury bill rates because either the real interest rate and/or expected inflation would increase depending on whether the central bank partially or fully monetizes budget deficits (Geske and Roll 1983).

Another explanation was given by Kaul (1987), who noted that the negative relationship between stock returns and expected inflation in the post-war period in the U.S. could be explained by counter-cyclical monetary policy. He suggested that positive shocks to real output generate monetary tightening and vice versa. There are also those who have argued that higher rates of inflation tend to be more variable. This variability in future inflation rates promotes uncertainty and depresses future economic activity (Hu and Willett 2000; Park and Ratti 2000).

Modigliani and Cohn (1979) also postulated that stock market investors suffer from money illusion by incorrectly discounting (inflation-unadjusted) equity cash flows with nominal (inflation-adjusted) discount rates. This leads to undervaluation in stocks when expected inflation is high and overvaluation when expected inflation is low. The inflation illusion hypothesis was tested by Lee (2010), who found that this theory holds in the post-war data, but could not explain a positive relationship found in prewar U.S. data.

The contradictory results in the empirical relationship between inflation and stock returns could be caused by the various models and specifications used in the testing methods of researchers. Kolluri and Wahab (2008) noted that many researchers used symmetric testing methods which gave consistent negative results, and that the methods used to calculate inflation could also have an effect on the results. They stated that the use of asymmetric testing methods can help distinguish differences between high and low inflation regimes and can provide a better picture of the true relationship.

Furthermore. Madsen (2007) compared the results based on the specification of the estimation equation, the measurement of inflation expectations, the time-series properties of inflation, and the time aggregations of the data. He showed that the higher the time-aggregation of the data and the better the instruments, the more likely it is that one finds evidence for the Fisher effect. The more persistent the inflation, the more likely it is that the Fisher effect is accepted using nominal share returns as the dependent variable.

Using monthly data, Alagidede and Panagiotidis (2010) studied the relationship between stock prices and consumer prices in six African countries (Egypt, Kenya, Morocco, Nigeria, South Africa, and Tunisia) where inflation has been persistent. Their results showed an overall positive relationship for both the short and long run, leading them to conclude that stocks were a hedge for inflation in these African countries.

A positive relationship was shown by Kim and Ryoo (2011), who used a two-regime threshold vector error-correction model on century-long U.S. data. They found that stocks were a hedge for inflation in U.S. since the 1950s. Omay et al. (2015) also confirmed a positive relationship between stock prices and the consumer price index using data from 52 countries, including the U.S., giving support to the generalized Fisher effect.

However, Kolluri and Wahab (2008) found an inverse relationship between stock returns and inflation forecasts during only low inflation periods, while a positive relationship was detected through high inflation periods in the United States. Overall they concluded that stocks provided favourable inflation protection (see Choudhry (2001); Li et al. (2010)).

These results from Kolluri and Wahab (2008) suggest that the relationship between expected inflation and stock returns depends somewhat on the country's inflation rate over the period. In an attempt to understand the role inflation rate and consumer price index play, this study proposes testing the Fisher effect in two groups of countries, the G7 group of low inflation industrialised countries (Canada, France, Germany, Italy, Japan, United Kingdom and United States) and the BRICS group of relatively high inflation emerging countries (Brazil. Russia, India, China, and South Africa).

Data and Model Estimation

Following Alagidede and Panagiotidis (2010), short-run regression models and long-run cointegration techniques are used for the analysis. The relationship between stock returns and inflation is estimated using available monthly data of consumer prices and stock prices from 1991:01 to 2014:12 for the G7 countries, India and South Africa, 1993:06 to 2014:12 for Brazil, 1997:09 to 2014:12 for Russia, and 1999:01 to 2014:12 for China. The consumer price index and the all-share index data were drawn from the Monetary and Financial Statistics database of the Organisation for Economic Cooperation and Development (OECD 2015).

The Fisher effect is extended to common stocks to explain the short- run relationship between stock returns and inflation.

[R.sub.t] = [E.sub.t-1] [r.sub.t] + [E.sub.t-1][[pi].sub.t] + [[mu].sub.t] (1)

where [R.sub.t] are the nominal returns, [E.sub.t - 1] [r.sub.t] are the expected real returns and [E.sub.t-1][[pi].sub.t] are the expected inflation rates. Assuming perfect foresight, inflationary expectations equal actual inflation and Equation. (1) becomes

[R.sub.t] = [E.sub.t-1] + [beta][[pi].sub.t] + [[mu].sub.t], (2)

For stocks to be a hedge against inflation in the short-run, [beta] [greater than or equal to] 1. In the presence of taxes, nominal stock returns are expected to rise by more than unity in response to expected inflation. Feldstein (1980) predicted the nominal returns to rise at a rate of 1/(1 - t) where t was the tax rate.

This article seeks to explain short-run relationship differences, if any, between G7 countries and the BRICS group of countries by extending the Fisher effect to common stocks using Equation. (2) above. Johansen cointegration tests and vector error correction methods are used to determine the long-run relationship between stock prices and consumer prices in the two groups of countries.

Assuming rational expectations implies zero mean error terms for Equations. (3) and (4) below:

[[pi].sub.t] = [E.sub.t-1][[pi].sub.t] + [[epsilon].sub.1t] (3)

[r.sub.t] = [E.sub.t-1] [r.sub.t] + [[epsilon].sub.2t]. (4)

Since only the nominal interest rate [R.sub.t] and the actual inflation [[pi].sub.t] are observable, the stationarity properties of [r.sub.t], the ex-post real interest rate is determined by the stationary properties of [R.sub.t] and tt;. Expressing equations (1) and (3) in terms of [E.sub.t-1] [r.sub.t] and [E.sub.t-1][[pi].sub.t] respectively and substituting into equation (4) gives equation (5) below:

[r.sub.t] = [R.sub.t]-[[pi].sub.t] + [[epsilon].sub.t] (5)

where [[epsilon].sub.t] = [[mu].sub.1t] - [[epsilon].sub.2t], and it is assumed to be stationary. If both [R.sub.t] and [[pi].sub.t] are stationary, then the real interest rate, [r.sub.t] is also stationary, since it is a linear combination of [R.sub.t] and [[pi].sub.t]. If both are of unit roots, then a cointegrating relationship may exist. This is the basis for testing the Fisher effect and its application to common stocks (Rapach and Weber 2004; Alagidede and Panagiotidis 2010).

For a long-run investigation, stock prices ([SP.sub.t]) and consumer prices ([CPI.sub.t]) are used instead of stock returns [R.sub.t] and inflation [[pi].sub.t] respectively. A standard unit root process as shown below is used in the analysis, where [y.sub.t] is a vector of [SP.sub.t] and [CPI.sub.t]:

[y.sub.t] = [delta] + [rho][y.sub.t-1] + [v.sub.t] (6)

[v.sub.t] is assumed to be an independent and identically distributed random variable with zero mean, and is independent of the observed initial value [y.sub.0] x [y.sub.t] has a unit root of, [rho] = 1. Therefore, the null hypothesis is that [y.sub.t] has a unit root against a one-sided alternative of [rho] < 1, and it is strictly -1 and 1.

For n series with unit roots and n-1 cointegrating vectors, a long-run relationship implies the series shares a common stochastic trend. Since fewer cointegration vectors mean fewer factors driving the variables towards equilibrium, one cointegrating vector (n-1 cointegrating vectors) implies a very stable long-run relationship between SP, and [CPI.sub.t]. The linear relationship, stock prices minus consumer prices, [SP.sub.t] - [CPI.sub.t] is stationary and also implies that the vectors are strongly cointegrated and have converged with a cointegrating vector of [1, -1]. A weaker and more general version of the vector of deviation from the long-run is given below;

[SP.sub.t-1]-[alpha]-[beta][CPI.sub.t-1]. (7)

This can be normalised and expressed as:

[SP.sub.t] = [alpha] + [beta][CPI.sub.t]. (8)

Stationarity is a necessary condition for cointegration but it is not sufficient as cointegration also requires that the elasticity of stock prices with respect to consumer prices be positive: [beta] > 0. A strong version requires that [alpha] = 0 and [beta] = 1, which implies a full hedge or a one-to-one relationship, since both variables are expressed in logarithms. Stock prices outperform consumer price if [beta] > 1.

Empirical Results

This section investigates the short-run and long-run relationship between short returns and consumer prices. Table 1 below shows the descriptive statistics of the G7 group of countries over the period 1991:01 to 2014:12. Average annual inflation means were positive: and stock returns were also positive, with the exception of Japan, but were more volatile, as measured by their standard deviations. The kurtosis for both series in the United Kingdom and United States, and inflation in Germany, were significantly higher than three, the level for a standard normal distribution. All the G7 group of countries have either adopted inflation targeting monetary policy or in the case of the U.S. Federal Reserve, the European Central Bank and the Bank of Japan, have many of the main elements of inflation targeting.

The descriptive statistics of the BRICS group of countries, on the other hand, indicates greater variability and higher positive average annual inflation means than the G7 group of countries, and many of the series show values that are not consistent with a standard normal distribution. Brazil and South Africa, the only BRICS countries with an inflation targeting monetary policy, have average inflation values of 30.05 and 6.29 respectively.

The study extends the Fisher effect to common stocks to explain the short-run relationship between stock returns and inflation in Table 2. Monthly inflation is calculated using CPI, data for each country as [[pi].sub.t] = 100(In[CPI.sub.t] - In[CPI.sub.t-12]) and monthly stock returns using [SP.sub.t] as [R.sub.t] = 100(In[SP.sub.t] - In[SP.sub.t-12]):

Nominal Stock Returns = [alpha] + [beta]Inflation + [[mu].sub.t]

For stocks to be a hedge against inflation in the short-run, [beta] [greater than or equal to] 1; [alpha] is the expected real returns.

The beta coefficients in Table 2 are significant in France, Italy, the U.S., Brazil, China, and South Africa. However, only the U.S., Brazil, and China, have positive and statistically significant slope estimates. The results suggest that a positive relationship does not depend on the country's inflation rate, and that stocks in the U.S., Brazil, and China are a hedge against inflation in the short run over the sample period as beta is greater than one., The results are an unreliable guide to the long-run relationship, as they appear to have been influenced by the high volatility of stock returns.

Regressing variables in first differences may lose important details about their long-run relationship so this study also uses the variables in levels to analyse long-run relationships. A long-run relationship, between stock prices and consumer prices requires us to know the integration and stationarity properties of the series. Stationarity tests in autoregressive time series models continue to receive considerable attention in econometric analysis. Ng-Perron tests are preferred because the Ng and Perron (2001) test statistic avoids the power problems usually associated with traditional methods for testing unit roots by putting together modified versions of Phillips and Perron (1988), Bhargava (1986), and the point optimal statistic by Elliott et al. (1996).

Tables 3 shows the results of the unit roots and stationarity tests, compared with critical values from Ng and Perron (2001), for the G7 and BRICS countries. The null hypothesis in the Ng-Perron test is that the time series is stationary. Test results for all the G7 and BRICS countries show that the variables are nonstationary in levels. The variables have to be differenced once to obtain stationarity. They are integrated of order one or I (1), except for the Italian consumer prices variable, which is integrated of order two or I (2). This suggests that shocks to the variables only have transitory effects and they will move back to equilibrium in the long run.

Starting with VAR (12), the Akaike information criterion and Schwarz information criterion are used to determine VAR lag order that best describes the data. The Johansen (1995) cointegration test is used to determine if the unit root or I (1) variables are cointegrated. The results of the test, specified without a trend, confirm that all the I (1) variables have a cointegrating equation at the 5% level. An ADF test on the residuals also indicates the presence of cointegration. Stability tests identify breaks in the cointegration relationship for Brazil between 1993 and 1995, and for Russia between late 1998 and 1999. Having established that the variables are cointegrated over the sample periods, long-run relationships are exploited using a vector error correction model (VECM). The variables in the long-run relationship are converted into natural logarithms so that an estimated coefficient indicates stock price elasticity.

The error correction terms in Table 4 represent the short-term response necessary to move the system back to long run equilibrium. In the long run this term is zero. However, when there is a deviation from the long run, the variables adjust to partially restore the equilibrium relationship.

The extension of the Fisher effect for stock returns implies that the levels of real stock prices are non-stationary, unpredictable, and cannot be modeled or forecasted, assuming a positive expected return. However, the results indicate a positive long-run relationship between stock prices and the consumer price index in both groups of countries, indicating that a positive relationship does not depend on the country's consumer price index. Italy was not included in the vector error correction model because its stock prices and consumer prices were not integrated of the same order. The estimated coefficients show the elasticity of changes in stock prices with a change in consumer price levels. For instance, a 1% change in Russian consumer price levels changes the Russian All-Share Index by 4.50%. Although the results of the short-run analysis for many of these countries were either not statistically significant or showed negative slope estimates, investors were fully compensated for consumer price fluctuations in the long run as stock prices outperformed consumer prices. The speed of return to the equilibrium position is fastest in Brazil, an error correction term of -0.048 indicates that 4.8% of the disequilibrium is corrected monthly. South Africa's error correction term is the slowest, less than 1% correction monthly to return the system to long-run equilibrium.

Conclusion

The Fisher effect applied to the stock market suggests that stock returns would move one-to-one with inflation, thus making stocks a hedge for inflation. Results of early empirical studies on the relationship between stock returns and inflation were mixed.

In an attempt to understand the relationship between stock returns and inflation, this study proposed testing the Fisher effect in two groups of countries, the G7 group of low inflation industrialized countries (Canada, France, Germany, Italy, Japan, United Kingdom and United States), and the BRICS group of relatively high inflation emerging countries (Brazil, Russia, India, China, and South Africa). The study then examines the extent to which a one-to-one movement of stock returns and inflation, in the two groups of countries, relates to the country's inflation rate in the short run.

The extension of Fisher effect to common stocks enabled an examination of the short-run relationship between stock returns and inflation. The study used cointegration and vector error correction models to investigate the long-run relationship between consumer price index and stock prices, and whether a positive relationship depends on the country's consumer price index.

In the short run, the US, Brazilian and Chinese stocks were a hedge against inflation over the sample period as their betas were greater than one., Although other countries such as, Canada, Germany, the UK, Russia, and India also showed positive relationships between stock returns and inflation, they were not significant and did not establish a one-to-one relationship. Results for France, Italy, and South Africa indicated a significant but negative relationship between stock returns and inflation, highlighting the mixed results from other studies. The evidence though, except for France, Italy, and South Africa, seems to be broadly consistent with a suggestion of a positive relationship at the country level, and does not depend on the country's inflation rate.

A limitation of the short-run analysis is that it throws away important information contained in the data because it uses stock returns and inflation, variables in first difference, rather than stock prices and the consumer price index. Over the long run, a 1% change in consumer price levels changes the Japanese All-Share Index by 10.81 % and the Chinese All-Share index by 1.05%. Therefore the study rejects the prior expectation that a positive Fisher relationship may depend on a country's inflation rate in the short-run and/or the consumer price index in the long-run.

The results of this short-run study do not support a relationship between a country's inflation rate and a one-to-one Fisher effect. The cointegration test supports the existence of a long-run relationship between stock prices and consumer prices in both the G7 group of low inflation industrialized countries (except Italy) and the BRICS group of relatively high inflation emerging countries. Although the results of the short-run analysis for some of these countries were not significant, investors were fully compensated for consumer price fluctuations in the long run, with Japan having the highest estimate and China being the lowest. The results have profound policy implications for investors because the findings suggests that neither the consumer price level nor the monetary policy regime alters the generalized Fisher effect.

Acknowledgements I would like to thank Rod Hill of the University of New Brunswick in Canada for his helpful comments and suggestions. The views presented here are solely those of the author and do not represent any institution.

DOI 10.1007/s11293-017-9543-9

Published online: 29 May 2017

References

Alagidede, P., & Panagiotidis. T. (2010). Can common stocks provide a hedge against inflation? Evidence from African countries. Review of Financial Economics, 19(3), 91-100.

Bhargava, A. (1986). On the theory of testing for unit root in observed time series. Review of Economic Studies. 53(3), 369-384.

Choudhry, T. (2001). Inflation and rates of return on stocks: Evidence from high inflation countries. Journal of International Financial Markets Institutions and Money, 11(1), 75-96.

Elliott, G., Rothenberg, T., & Stock, J. K. (1996). Efficient tests for an autorcgrcssivc unit root. Econometrica, 64(4), 813-836.

Fama, E. F. (1981). Stock returns, real activity, inflation, and money. American Economic Review, 71(4), 545565.

Feldstcin, M. (1980). Inflation and the stock market. American Economic Review, 70(5), 839 847.

Fisher, I. (1930). The theory of interest (pp. 21-27). New York: Macmillan.

Gallagher, L. A.. & Taylor. M. P. (2002). The stock return-inflation puzzle revisited. Economic Letters, 75(2), 147-156.

Gcske, R., & Roll, R. (1983). The fiscal and monetary linkage between stock returns and inflation. The Journal of Finance, 35(1), 1-33.

H[u.sub.i] X., & Willctt, T. (2000). The variability of inflation and real stock returns. Applied Financial Economics, 10(6), 655-665.

Johansen, S. (1995). Likelihood-based inference in cointegrated vector autoregressive models (pp. 70-106). Oxford: Oxford University Press.

Kaul, G. (1987). Stock returns and inflation: The role of the monetary sector. Journal of Financial Economics, 18(2), 253-276.

Kim, J. H., & Ryoo, H. H. (2011). Common stocks as a hedge against inflation: Evidence from ccntury-long US data. Economic Letters, 113(2), 168-171.

Kolluri, B., & Wahab, M. (2008). Stock returns and expected inflation: Evidence from an asymmetric test specification. Review of Quantitative Finance and Accounting, 50(4), 371-395.

Lee, B. S. (2010). Stock returns and inflation revisited: An evaluation of the inflation illusion hypothesis. Journal of Banking & Finance, 34(6), 1257-1273.

Li, L., Narayan, P. K., & Zheng, X. (2010). An analysis of inflation and stock returns for the UK. Journal of International Financial Markets Institutions and Money, 20(5), 519-532.

Madsen, J. (2007). Pitfalls in estimates of the relationship between stock returns and inflation. Empirical Economics, 35(1), 1-21.

Modigliani, F., & Cohn, R, A. (1979). Inflation, rational valuation and the market. Financial Analysts Journal 35(2), 22-44.

Ng, S., & Perron, P. (2001). Lag selection and the construction of unit root tests with a good size and power. Econometrica, 69(6), 1519-1554.

OECD (2015). Monthly Monetary and Financial Statistics. Organization for Economic Cooperation and development, accessed at: https://stats.occd.org/indcx.aspx?queryid=6779, May 1, 2015.

Omay, T., Yuksel, A., & Yuksel, A. (2015). An empirical examination of the generalized Fisher effect using cross-sectional correction robust tests for panel cointegration. Journal of International Financial Markets, Institutions & Money, 35(1), 18-29.

Park, K., & Ratti, R. (2000). Real activity, inflation, stock returns, and monetary policy. The Financial Review, 35(2), 59-78.

Phillips, P. C. B., & Perron, P. (1988). Testing for a unit root in time series regression. Biometrika, 75(2), 335346.

Rapach, D. E., & Weber, C. E. (2004). Are real interest rates nonstationary? New evidence from tests with good size and power. Journal of Macroeconomics, 26(3), 409-430.

Mustapha Ibn Boamah (1)

[mail] Mustapha Ibn Boamah mboamah@unb.ca

[1] Faculty of Business, University of New Brunswick, Saint John. NB, Canada
Table 1 Descriptive statistics of G7 countries

                  Variable     12-month             Std. Dev.
                               Average (percent)
G7 Countries

  Canada          Inflation    1.798                0.889
                  Returns      6.280                17.050
  France          Inflation    1.576                0.727
                  Returns      4.681                21.590
  Germany         Inflation    1.819                1.121
                  Returns      4.717                23.217
  Italy           Inflation    2.582                1.301
                  Returns      3.023                24.922
  Japan           Inflation    0.224                1.085
                  Returns      -1.638               22.761
  United Kingdom  Inflation    2.219                1.048
                  Returns      4.431                15.058
  United States   Inflation    2.405                1.067
                  Returns      6.901                16.699
BRICS Countries

  Brazil          Inflation    30.049               80.202
                  Returns      24.669               65.687
  Russia          Inflation    31.193               46.699
                  Returns      19.017               46.961
  India           Inflation    7.311                3.166
                  Returns      12.329               30.744
  China           Inflation    3.803                5.600
                  Returns      3.317                37.642
  South Africa    Inflation    6.296                3.191
                  Returns      10.889               18.053

                    Skewness   Kurtosis

G7 Countries

  Canada            -0.064     3.709
                    -0.814     3.821
  France            -0.377     3.309
                    -0.778     3.060
  Germany            1.518     5.992
                    -0.949     3.489
  Italy              0.483     2.899
                    -0.367     3.228
  Japan              0.724     3.624
                     0.066     2.563
  United Kingdom     1.372     6.140
                    -1.031     4.007
  United States     -0.836     5.590
                    -1.625     6.921
BRICS Countries

  Brazil             3.412     13.280
                     3.772     20.795
  Russia             2.512      8.839
                    -0.259      4.518
  India              0.379      2.879
                     0.093      3.490
  China              2.068      7.014
                     0.447      4.671
  South Africa       0.063      3.479
                    -0.962      4.345

Source: Own computation using monthly data from 1991:01 to 2014:12
for the G7 countries, India and South Africa, and 1993:06 to
2014:12 for Brazil, 1997:09 to 2014:12 for Russia, 1999:01 to
2014:12 for China; from the OECD (2015)

Table 2 Short-run analysis

                  [alpha]          [beta]

G7 Countries
  Canada          3.377 (2.313)    1.615 (1.154)
  France          13.259 (3.061)   -5.444 * (1.765)
  Germany         3.952 (2.671)    0.421 (1.251)
  Italy           13.157 (3.274)   -3.925 * (1.133)
  Japan           -1.612 (1.402)   -0.115 (1.268)
  UK              2.884 (2.127)    0.656 (0.867)
  US              -2.479 (2.409)   3.899 * (0.916)
BRIC Countries
  Brazil          5.478 (2.080)    1.046 * (0.036)
  Russia          13.352 (4.723)   0.378 (0.223)
  India           11.949 (4.672)   0.052 (0.587)
  China           -4.175 (3.960)   3.307 * (1.254)
  South Africa    23.297 (2.260)   -1.971 * (0.320)

                  [R.sup.2]    DW

G7 Countries
  Canada          0.007        0.136
  France          0.034        0.096
  Germany         0.001        0.086
  Italy           0.042        0.102
  Japan           0.000        0.097
  UK              0.002        0.114
  US              0.062        0.102
BRIC Countries
  Brazil          0.778        0.167
  Russia          0.015        0.104
  India           0.000        0.112
  China           0.038        0.070
  South Africa    0.121        0.147

Source: Own computation using monthly data from 1991:01 to 2014:12
for the G7 countries, India and South Africa, 1993:06 to 2014:12
for Brazil, 1997:09 to 2014:12 for Russia, and 1999:01 to 2014:12
for China, from the OECD (2015). indicates statistical significance
at 5%. Standard errors are in parentheses.

Table 3 Ng-Perron unit root tests (Exogenous:
Constant; Lag length = 1) (a)

                       [MZ.sub.[alpha]]     [MZ.sub.t]

G7 Countries
  Canada
    Stock prices       0.724 (-55.836 *)    0.718 (-5.265 *)
    Consumer prices    1.475 (-107.702 *)   4.280 (-7.265 *)
  France
    Stock prices       0.047 (-37.147 *)    0.037 (-4.292 *)
    Consumer prices    1.400 (-153.634 *)   5.588 (-8.761 *)
  Germany
    Stock prices       -0.116 (-47.804 *)   -0.069 (-4.886 *)
    Consumer prices    1.292 (-66.156 *)    5.395 (-5.731 *)
  Italy
    Stock prices       -0.923 (-49.717 *)   -0.583 (-4.966 *)
    Consumer prices    1.172 (-5.708)       5.091 (-1.574)
  Japan
    Stock prices       -4.096 (-25.368 *)   -1.419 (-3.548 *)
    Consumer prices    -0.619 (-104.597 *)  -0.332 (-7.221 *)
  United Kingdom
    Stock prices       0.439 (-27.663 *)    0.439 (-3.678 *)
    Consumer prices    1.618 (-62.263 *)    5.814 (-5.557 *)
  United States
    Stock prices       0.868 (-17.549 *)    1.069 (-2.914 *)
    Consumer prices    1.314 (-176.147 *)   3.492 (-9.327 *)
BRICS Countries
  Brazil
    Stock prices       0.448 (-87.689 *)    0.638 (-6.615 *)
    Consumer prices    0.488 (-17.578 *)    0.818 (-2.964 *)
  Russia
    Stock prices       0.069 (-63.473 *)    0.055 (-5.625 *)
    Consumer prices    0.751 (-61.998 *)    2.279 (-5.566 *)
  India
    Stock prices       1.064 (-53.743 *)    1.111 (-5.159 *)
    Consumer prices    1.669 (-92.942 *)    5.601 (-6.817 *)
  China
    Stock prices       -2.182 (-50.664 *)   -0.853 (-4.829 *)
    Consumer prices    0.969 (-147.767 *)   1.905 (-8.594 *)
  South Africa
    Stock prices       1.227 (-116.124 *)   1.604 (-7.600 *)
    Consumer prices    1.358 (-42.383 *)    5.344 (-4.528 *)

                       MSB                [MP.sub.T]

G7 Countries
  Canada
    Stock prices       0.992 (0.094 *)    65.197 (0.486 *)
    Consumer prices    2.902 (0.067 *)    593.041 (0.361 *)
  France
    Stock prices       0.783 (0.116 *)    37.401 (0.712 *)
    Consumer prices    3.990 (0.057 *)    1098.27 (0.165 *)
  Germany
    Stock prices       0.595 (0.102 *)    23.738 (0.519 *)
    Consumer prices    4.177 (0.087 *)    1176.49 (0.417 *)
  Italy
    Stock prices       0.632 (0.099 *)    21.393 (0.543 *)
    Consumer prices    4.344 (0.276)      1240.10(4.646)
  Japan
    Stock prices       0.347 (0.139 *)    5.998 (1.013 *)
    Consumer prices    0.537 (0.069 *)    18.605 (0.254 *)
  United Kingdom
    Stock prices       1.000(0.133 *)     62.137 (1.019 *)
    Consumer prices    3.593 (0.089 *)    931.263 (0.447 *)
  United States
    Stock prices       1.231 (0.166 *)    99.621 (1.575 *)
    Consumer prices    2.659 (0.053 *)    482.827 (0.223 *)
BRICS Countries
  Brazil
    Stock prices       1.425 (0.075 *)    119.126 (0.293 *)
    Consumer prices    1.677 (0.169 *)    163.865 (1.394 *)
  Russia
    Stock prices       0.799 (0.088 *)    38.865 (0.405 *)
    Consumer prices    3.032 (0.089 *)    553.876 (0.400 *)
  India
    Stock prices       1.043 (0.096 *)    76.486 (0.517 *)
    Consumer prices    3.356 (0.073 *)    821.171 (0.264 *)
  China
    Stock prices       0.391 (0.095 *)    9.811 (0.997 *)
    Consumer prices    1.964 (0.058 *)    248.074(0.168 *)
  South Africa
    Stock prices       1.307 (0.065 *)    119.868 (0.246 *)
    Consumer prices    3.935 (0.107 *)    1058.92 (0.787 *)

Source: Own computation using monthly data from 1991:01 to 2014:12
for the G7 countries. India and South Africa. 1993:06 to 2014:12
for Brazil. 1997:09 to 2014:12 for Russia, and 1999:01 to 2014:12
for China. From the OECD (2015). * Accept the null hypothesis of
stationarity in first difference at 5%.

* Values in parenthesis are for the variables in difference. All
the Ng-Perron tests reject the null hypothesis of stationarity at
5% in levels.

Table 4 Long-run analysis

                  Cointegrating Equation

G7 Countries
  Canada          [SP.sub.t] = 9.689 + [3.106.sub.(0.444)][CPI.sub.t]
  France          [SP.sub.t] = 0.837 + [1.177.sub.(1.591)][CPI.sub.t]
  Germany         [SP.sub.t] = 1.541 + [1.335.sub.(0.794)][CPI.sub.t]
  Japan           [SP.sub.t] = 44.957 + [10.807.sub.(4.663)][CPI.sub.t]
  UK              [SP.sub.t] = 2.406 + [1.542.sub.(0.381)][CPI.sub.t]
  US              [SP.sub.t] = 5.591 + [2.247.sub.(0.575)][CPI.sub.t]
BRICS Countries
  Brazil          [SP.sub.t] = 4.198 + [1.809.sub.(0.159)][CPI.sub.t]
  Russia          [SP.sub.t] = 14.802 + [4.503.sub.(0.679)][CPI.sub.t]
  India           [SP.sub.t] = 2.995 + [1.596.sub.(0.240)][CPI.sub.t]
  China           [SP.sub.t] = 0.491 + [1.053.sub.(0.783)][CPI.sub.t]
  South Africa    [SP.sub.t]= 5.533 + [2.241.sub.(0.109)][CPI.sub.t]

                  Error Correction Term

G7 Countries
  Canada          [-0.032.sub.(0.015)]
  France          [-0.015.sub.(0.008)]
  Germany         [-0.038.sub.(0.012)]
  Japan           [-0.021.sub.(0.011)]
  UK              [-0.037.sub.0.012)]
  US              [-0.018.sub.(0.011)]
BRICS Countries
  Brazil          [-0.048.sub.(0.016)]
  Russia          [-0.009.sub.(0.004)]
  India           [-0.027.sub.(0.012)]
  China           [-0.034.sub.(0.015)]
  South Africa    [-0.008.sub.(0.010)]

Source: Own computation using monthly data from 1991:01 to 2014:12
for the G7 countries, India and South Africa, 1993:06 to 2014:12
for Brazil, 1997:09 to 2014:12 for Russia, and 1999:01 to 2014:12
for China. From the OECD (2015). Standard errors are in
parentheses.
COPYRIGHT 2017 Atlantic Economic Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Boamah, Mustapha Ibn
Publication:Atlantic Economic Journal
Article Type:Report
Geographic Code:9INDI
Date:Jun 1, 2017
Words:5627
Previous Article:Do #AllLivesMatter? An evaluation of race and excessive use of force by police.
Next Article:California's three strikes law revisited: assessing the long-term effects of the law.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters