In stock market prediction, there are many contradicting ideas and assertion that demand investors to take serious attention and to perform scientific research to falsify the false assertion. For example, just to name a few complication and seemingly contradictory opposite assertion in the context of investing and stock market prediction, the following issues are field being hotly debated by scholars and practitioners: (a) either if the stock market is efficient or not, (b) if Price to Earnings ratio can be used to predict stock returns, (c) if there are truly investment methods that can be used effectively to outperform the market over the long term, (d) if technical analysis can be employed for superior profitability in the financial market, and many others. Indeed, such contradictory assertions are not only causing great confusion to the public, but also to experienced investors in the financial markets. However, for any investor wishing to obtain superior performance from their investment, it is crucial to identify the wrong conceptions popular among the investment fraternity. This is because false concepts will often detrimental to the wealth and performance of an investors – and can be deadly when the investor believe in these wrong conceptions faithfully without questioning the validity of the false assertion. In this dissertation, the seemingly myth, that economic indicators can be used to predict stock returns will be investigated. It is strongly believed that truth seeking is crucial towards investment success, and through the application of scientific methods, the assertions of different parties in the context of macroeconomic indicators and stock returns will be scrutinized.
Any students or participants of the financial market will likely to be familiar with the theoretical concepts of macroeconomic and how the various forces in macroeconomic may affect the financial market. This is not surprising, as the news of macroeconomic is constantly being published in investment related books, newsletter, journals, literature and even investment textbook. Apparently, macroeconomic is not something can be ignored, given its relative importance and macroeconomy as a highly popular topic in frequently discussed among mass media.
The notion that macroeconomic indicators are useful in predicting stock returns is closely related to the concept of business cycles. For this, according to the concept of business cycle, it is perceived that the economic health of a society or nation is often cyclical. In different stages of the business cycle, macroeconomic indicators will tend to portray different picture of a country economy. In the context of business cycle, among the popular macroeconomic variables frequently discussed include the following: national output, interest rates, inflation rates, money supply, unemployment situation, consumer confidences, as well as corporate profits (Reilly & Brown, 2003). Indeed, there are so many macroeconomic variables being published from time to time. The macroeconomic variables are mentioned purely due to their popularity among investment fraternity.
The notion that macroeconomic variables or indicators should be able to predict stock returns is not something irrational. Firstly, as business cycles changes, the respective macroeconomic variables will also change accordingly. Many of the macroeconomic variables have significant impacts towards corporate profitability. For example, macroeconomic variables such as inflation rate, interest rates or the rate of growth in gross domestic products are closely related and indeed, may be influential towards the profitability of corporate profits. From another angle, stock returns are closely related to the corporate profits and earning prospects. When the macroeconomic variables changes, it is reasonable to expect that corporate profits may change accordingly. When these happen, stock prices will change to reflect the changes in the corporate profitability or earning prospects. As such, it is indeed reasonable to expect that there should be certain degree of relationships between macroeconomic indicators or variables to stock returns, given the close linkages between business cycles, macroeconomic variables and corporate profitability (Maginn et. al., 2007). From such perspective, proponents of macroeconomic indicators can be used to improve investment results is based on the idea that any investors able to discern the change of macroeconomic picture will be able to yield superior returns by switching out from the stock market before recession set in, and then to allocate capital into the stock market when the economy is expected to recover (Tainer, 2006). Indeed, according to the computation by Siegal (2002), if any investors able to switch correctly from stock to bond market and vice versa before the turning of economy tide, he will be able to realized an excess return of 4.8% per annum when compared to the buy and hold strategy. To be fair, an excess return of 4.8% per annum is indeed significant, and deserves serious and immediate attention by serious investors. This is perhaps one of the reasons there are many studies performed to understand the relationships between economic indicators to stock returns. Secondly, it is sensible to believe that there are certain degree of relationships between stock returns and economic indicator today. This is because a review of the literature (as will be further articulated in-depth in the following sections), there are some studies able to discover statistically significant relationships between economic variables to stock returns. Considering these two reasons, it should indeed provide significant confidence to any investors to spent time to understand the interaction between economic variables to stock market.
There are certain macroeconomic indicators that are given more importance by the practitioners or academicians in investigating stock market. This is probably due to the importance of these macroeconomic variables in affecting the turns of business cycles and corporate profits. In this section, the different macroeconomic variables and their impacts to business cycle and stock returns will be discussed.
In the business cycle model portrayed by Tainer (2006) and Baumohl (2005), business cycle can be characterized as having five distinctive stages. The first stage is about recession. In this first stage, the growth of national output may decrease, and unemployment rate may rise. Technically speaking, recession is defined as a situation whereby there are two back-to-back quarters of negative Gross Domestic Products growth in a nation (Moss, 2007). Then, the second stage of the business cycle is called the trough, which is technically defined as the minimum point of a business cycle. In such stage, inflation rates are low. Unemployment rate is likely to be high. Business activities are slow, and corporate profits will likely to be unsatisfactory. Weaker firms may suffer badly from serious financial distress or face the fate of bankruptcy. It is often within such stage that the central bank of a country will implement an expansionary monetary policy to simulate the economy (Tainer, 2006; Bodie, Kane, & Marcus, 2007). The main measure to stimulate the economy is through increasing the money supply and lowering of interest rates to encourage investment and spending. When that happens, corporations are likely to more willingly expand or invest; this will in turn create jobs, revive consumer confidences and spending and subsequently reviving the business outlook in a country. When expansionary monetary policy is successful, the business cycle enters into the third stage. Third stage is called recovery. Some of the economic indicators (related to national output and growth of gross domestic products) are likely to improve. In the recovery stage, unemployment rate may decrease, and inflation rates may increase slowly but marginally (Moss, 2007). Besides, corporate profitability may be restored, and business prospects may simply improve as a result. Technically speaking, recovery process is perceived to have ended, taken over by the fourth stages, whereby the output loss in the previous recession is recuperated. The end of the recovery stage will be replaced by the expansion stage. In the expansion stage, business outlook become increasingly vibrant and inflationary pressures set in. As discussed by Reilly & Brown (2003), central banks around the world today seemingly have more interests in stabilizing the economy, and are adopting an inflation targeting approach governed the price level. As such, at the end of the expansion stage, central banks are likely to increase interest rates to combat the increasingly threatening inflationary pressure. This means central bank is implementing a restrictionary monetary policy, to reduce the growth of money supply in the economy. Often, interest rates will be increased as long as inflationary pressures are haunting the economy. Lastly, expansion stage will continue until it reach the fifth stage, namely the peak, whereby a marginal increase in interest rates will turn the stock market bearish and the economy into recessionary (Tainer, 2006).
A review of the changes of the respective economic variables in the different stages of a business cycle indicate that certain economy variables are used to describe the different stages. These variables include, but not limited to the following: real economic activity (as proxy by gross domestic products), money supply, inflation rates, and interest rates and unemployment rate. The changes of these economic variables under the five stages of business cycles will be summarized in Table 2.1 below.
As discussed before, it is easily observed that macroeconomic variables have been employed by scholars and researchers in studying and investigating stock returns in financial market. There are many different macroeconomic variables being employed by researchers to study the relationships between these macroeconomic variables to stock returns. Among these macroeconomic, as discussed in great depth previously, include: real economic activity (as proxy by gross domestic products), money supply, inflation rates, and interest rates and unemployment rate. These variables are popularly selected due to the linkages of these variables to business cycles and economic activity, which has huge impacts against companies’ profitability, and hence, stock prices in the financial market. However, the impacts of these variables to stock returns may vary, dependent on the context and degree of influence of these variables towards the economy and profitability of listed corporations in a particular country. In the following paragraphs, several popular and widely discussed macroeconomic variables and the respective linkages to stock returns will be discussed. Empirical evidences on the relationships of these macroeconomic variables to stock returns will also be outlined.
The linkages of real economic activities and stock returns are intuitive – in the sense that real economic activities is ultimately about economic weaknesses or strengths of a country or the world, that will ultimately affect the future prospect or financial performance of businesses. There are various different economic variables, or indicators, used to proxy real economic activities. Among two of the most popular proxy for real economic activities are Gross Domestic Products and Industrial Production Index. In this section, the linkages of Gross Domestic Products (GDP) to stock returns will be discussed.
According to Tainer (2006), GDP is the most comprehensive figure on the vitality of the economy. GDP measures the speed of economic growth for a particular country. It is perceived as able to reflect the up and down of an economy – and thus, it is then not surprising that many economist or investors are focusing on GDP to judge the health of the economy, and then in turn to predict the stock market in the future. As discussed by Baumohl (2005), the health of economy is influential towards business activities and hence corporate profits. In a booming economy, businesses able to achieve higher earnings, and hence, often results in higher stock prices. Then, in a weak or gloomy economy, businesses often will face harder situation to generate earnings, and hence, such a situation often resulted in lower stock prices. However, as discussed by Tainer (2005), the drawback of using GDP as predictor of stock returns is that GDP is only available at quarterly basis. In the fast moving business environment, stock prices tend to move fast ahead of GDP. Often, forecaster would rely on monthly data more, as when the release of GDP for the past figure is available to the market, the information is likely to be already discounted in stock prices. As such, often, Industrial Production Index (IPI) is often used to predict stock returns. The relationships between IPI to stock returns will be discussed.
As discussed by Abdullah and Hayworth (1993), Industrial Production Index (IPI) is being released monthly in most of the countries, and it covers basically everything being produced physically in a nation. As discussed by Tainer (2006), one of the reason giving rise to the popularity of IPI is that this indicator often able to react fast to the turns f business cycle. Economist such as Baumohl (2005) agrees on such notion. As discussed by Baumohl (2005), this indicator is useful even in United State, whereby the large portion of the economy is relying on the services sectors. Accordingly, manufacturing activities are sensitive to changes in business cycle, while service sector tend to be more resilient to changes in economic climates, and hence, IPI can be effectively used to gauge the economic climate in a nation.
There is abundance of empirical evidences available concerning relationships between IPI and stock returns. However, it is noted that not all of the literature available discover statistically significant relationships between IPI and stock returns. In the following paragraphs, the studies discovering statistically significant relationships between IPI to stock returns will be firstly presented, followed by those studies failed to discover statistically significant relationship between the two variables.
By employing autoregressive bivariate correlation and multivariate regression concepts, Chen, Roll & Ross (1986) discovered that IPI was statistically priced in explaining expected stock returns in United States, from year 1953 to 1983. Similarly, Fama (1990) employed distributed lag regression and multiple regression techniques to investigate the relationships between stock returns to IPI in United States. It is discovered that future growth rate of IPI able to explain a total of 43% of the variances of annual stock returns. The research period for this study is from 1953 to 1987. Then, Chen (1991) documented that lagged IPI growth rate is one of the important macroeconomic determinant of stock returns in United States, from 1954 to 1984. Later, study by McGowan and Dobson (1993) also discovered that the rate of change in IPI is statistically significant in explaining industry returns in United States. However, it is apparent that stock returns lead IPI. A study by Gallinger (1994) found strong empirical evidences supporting causation from real stock returns to real economic activities, proxy by IPI, while only weak evidences supporting causation from IPI to stock returns in United States. Overall, prior to year 1970, there is abundance of literatures discovering statistical significant relationships between stock returns to IPI in United States. However, from year 1970 to 2000, a study conducted by Hamori, Anderson & Hamori (2002) found contradicting evidences against the notion that stock returns are a significant predictor of future real economic activities in United States. These scholars however, found and suggested that there were significant differences in the relationships between stock returns and IPI growth rates across time and culture (by comparing United States and Japan).
Next, the empirical evidences on relationships between stock returns to IPI of the respective country are mixed, whereby no statistically significant relationships between stock returns to IPI can be found. Some of the studies discovering statistical significant relationships between stock returns to IPI are as follow. By employing interbattery factor analysis, Bodurtha, Cho and Senbet (1989) found statistically significant evidences supporting the notion that IPI is able to explain the cross sectional average stock returns in the following countries: United States, Canada, United Kingdom, France, Germany, Australia and Japan. The period being investigated is from year 1973 to year 1983. Similarly, Lovatt & Parikh found that stock returns do indeed having statistically significant relationships with IPI in United Kingdom, from year 1980 to 1994. However, as discovered by Muradoglu, Taskin & Bigan (2000), out of a total of nineteen emerging countries around the world, there is no empirical evidence supporting that IPI Granger-cause stock return. Then, among these 19 emerging countries, only two countries, namely India and Mexico, have evidences that stock returns Granger-cause IPI. Then, similarly, employing VAR, cointegration, impulse response and variance decomposition methods, from year 1984 to 1999, Hondroyiannis & Papapetrou (2001) found that stock returns do not lead changes in real economic activities in Greece. Overall, this suggests that statistically significant between stock returns and IPI may not exist in every part of the world, and worst, such a relationship may alter from time to time.
In the financial market, monetary policy implemented by Central Bank has powerful influences against future and expected stock returns. As discussed by Bodie, Kane, & Marcus (2007), money supply is one of the most crucial indicators in relation to the monetary policy by central bank of a country. Indeed, there are successful practitioners or fund managers relying on money supply, or information related to monetary policies of central bank to make their investment decision. For example, Zweig (1997) is one of the famous fund managers relying on information related to monetary policies in stock market investment. Accordingly, Zweig (1997) asserted that in the stock market, it is similar to the horse racing events, whereby money is the driver that makes the mare run fast. Scholars agreed on the existence of relationships between money supply towards stock returns. For instance, Kim & Wu (1987) found that the money supply is one of the very significant priced risk factor in explaining stock returns in United States (year 1973-1985). Other scholars with similar opinion include the following: Flannery & Protopapadakis (2002), and Fifield, Power, Sinclair (2002).
The literatures concerning stock returns relationships to money supply are never limited. For example, Kraft & Kraft (1977) found empirical results indicating statistical relationships between money supply and stock prices in United States (year 1955:1 – 1974:12). However, according to the scholars, there is no causality running from either the money supply or percentage changes in the money supply to the level or percentage change in stock prices. Unidirectional causality exists from the stock price measures to money supply and percentage changes in money supply. Then, according to study by Pearce & Roley (1985), there are empirical evidences indicating unexpected surprises related to monetary policy tend to significantly affect stock prices (year 1977:9 – 1982:10). They discovered that the anticipated components of economic announcements, however do not significantly affect daily stock price movements in United States. In a similar manner, Abdullah & Hayworth (1993) found that stock returns are positively related to money growth (year 1980:4 – 1988:9). Indeed, they also found evidences supporting the notion that money growth Granger-cause stock return (i.e., the argument by Zweig (1997) that money makes the mares go in stock market). In a nutshell, findings of Abdullah & Hayworth (1993) suggest that money growth can explain a substantial proportion of the forecast error variance of stock returns. Their findings are also supported by study of Thorbecke & Coppock (1996). Thorbecke & Coppock (1996) found that on average, 32% of the variation in stock returns can be explained by monetary policy (year 1974:9 – 1979:9, 1982:8 – 1987:9). Specifically, news of contractionary monetary policy is found to be likely to trigger a large and statistically significant decline in stock returns. Besides, Park & Ratti (2000) also found that contractionary monetary policy shocks generate statistically significant movements in inflation and expected real stock returns, in which these movements head in an opposite direction in United States (year 1979-1982). Last but not least, Conover, Jensen, Johnson, & Mercer (2005) found empirical evidences supporting the idea that monetary policy continues to have a strong relationships with security returns, whereby stock returns are consistently higher and less volatile (i.e., less risky) when the Fed is promoting an expansive monetary policy (year 1963-2001).
In the international context, there are also empirical evidences supporting statistically significant relationships between money supply and stock returns. For instance, Kwon, Shin & Bacon (1997) found that money supply is one of the significant factors being priced in stock return in Korean stock market. To articulate, it is discovered that stock price indices are cointegrated (i.e., a long-run equilibrium relation) with money supply in Korean stock market (year 1980:1 – 1992:12). In Japan (as well as in United States), Friedman (2005) reported that monetary policy played a supporting role in various booms (year 1920s-2000s). Then, Patra & Poshakwale (2006) also discovered empirical results indicating that short run and long run equilibrium relationships exists between money supply and stock prices in Athens stock exchange (year 1990-1999).
Any discerning observers or investors will likely to notice that changes in interest rates are influential towards stock prices. Indeed, expectations on changes of interest rates often, are already powerful to shake the financial market. Common sense would justify the importance of interest rates towards stock returns. For example, any investors would likely consider investing his money in the stock markets or in the bond market (or perhaps, simply to save it in the banks to earn interests). For example, when interest rates increase, it looks attractive to buy bond or to save more fund to bank accounts. Thus, such a situation is likely to put pressures to the stock markets, whereby more conservative investors may simply sell the stocks they are holding (perhaps partially), to be saved in the bank. As such, stock prices are likely to be depressed. Besides, as argued by Siegal (2002), the rise of interest rates will also likely to affect corporate operations adversely. Rising interest rates indicate rising costs of doing business. For instance, rising interest rates can means rising interest payment and expenses for corporation with debt in the capital structure. Then, rising interest rates will also affect market sentiment (Yamarone, 2007), whereby corporations may delay or give up merger and acquisition activities for growth.
Many of the existing literature uncovered statistically significant relationships between interest rates and stock returns. For example, Kraft & Kraft (1977) found unidirectional causality exists from the stock price measures to interest rates (represented by Moody’s AAA corporate bond rate) in United States, employing data during period from 1955:1 – 1974:12. Then, Chen, Roll & Ross (1986) also found that changes in risk premia innovations are significantly priced in explaining expected stock returns (remark: risk premia is calculated as the spread between high grade and low grade bond) in United States (within 1953-1983). Apart from that, Chen, Roll & Ross (1986) also uncovered that twist on the yield curve innovations are significantly priced in explaining expected stock returns (remark: yield curve twist is calculated as the spread between long term and short term interest rates) in United States. Similarly, Kim & Wu (1987) found that the interest rates is one of the very significant priced risk variables in explaining stock returns in United States (research period within 1973-1985). Then, through univariate regression & multiple regressions, a study by Chen (1991) documented that the default spread (i.e., the difference between composite bond yield and high grade bond yield) is one of the important determinants of future stock market returns (year 1954-1986). Not only is that, Chen (1991) also documented that the term spread (i.e., difference between long term T-bond and short term T-bill) as well as 1-month T-bill rate are two of the important determinants of future stock market returns. Consistent with these literatures, Abdullah & Hayworth (1993) also discovered that short and long-term interest rates is found to Granger-cause stock returns and explain a substantial proportion of the forecast error variance of stock returns. In simpler words, they suggested that stock returns are negatively related to both short and long term interest rates. Similarly, McGowan & Dobson (1993) employed factor analysis & canonical correlation method, can found that term structure of the interest rates is statistically and economically significant in explaining industry returns. Besides, they also uncovered that default premium is statistically and economically significant in explaining industry returns. Perhaps most encouraging for aggressive and ambitious investors, findings from Resnick & Shoesmith (2002) is most encouraging. They found that the value of yield spread between 10-year T-bond and 3-month T-bill possesses crucial information about the probability of a bear market. By employing a probit model (from year 1960:1 – 1999:12), they documented that market timing based on yield curve could beat the buy-&-hold strategy. This strongly suggests that serious investors should never ignore the impacts of interest rates and yield curve towards future stock returns.
There are also empirical evidences supporting existence of statistically significant relationships between interest rates and stock returns in the international context, although the empirical evidences may not be conclusive. For example, Bodurtha, Cho & Senbet (1989) discovered that international bond returns is significant in explaining the cross-section of average stock returns in Canada, United Kingdom, France, Germany, Australia, and Japan (employing interbattery factor analysis, from year 1973:1 – 1983:12). Then, Hondroyiannis & Papapetrou (2001) discovered that impulse response analysis demonstrated that interest rates are important in explaining stock price movement (although a substantial proportion of the stock returns variations remain unexplained) in Greece. Specifically, real stock returns respond negatively to interest rates shocks (from year 1984:1 – 1999:9). Next, Fifield, Power, Sinclair (2002) also uncovered that interest rates are one of the economic factors able to explain returns in 13 emerging stock markets, using regression analysis from year 1987-1996. On the contrary, literature concerning absence of statistical relationships between interest rates and stock returns in the international context is never lacking. For instance, Kwon, Shin & Bacon (1997) found that interest rate related variables are not priced in stock return in Korean stock market (from year 1980:1 – 1992:12). Then, according to Muradoglu, Taskin, & Bigan (2000), the documented that out of 19 emerging countries, there are only evidences showing interest rates Granger-cause stock returns in Brazil, Pakistan and Zimbabwe (with research period ranging from 1976-1997).
Perhaps the relationships of interest rates and stock returns may differ accordingly to the different stages of business cycle or market sentiment at a particular point of time. Some researchers are studying the relationships of interest rates and stock returns with such assumption. Empirical evidences are available in supporting such idea. For instance, according to Ferson & Harvey (1991a), it is documented that compensation for bearing real interest rates risk is highest near business cycle troughs, from year 1964-1986. Then, as argued by Bolten & Weigand (1998), it is commented that changes in interest rates throughout the economic cycle are shown to cause changes in the level of stock prices, thus suggesting that monitoring and forecasting interest rates could help forecasting of stock prices behavior over time. Last but not least, in the recent year (from 1997-2002), a study by Funke & Matsuda (2006) suggested that there exist asymmetric reactions of stock prices to macroeconomic news. Precisely, in a booming (recession) economy, lower (higher) than expected interest rates may be good news for stock prices.
Inflation rate, or sometimes, referred to as the aggregate price level, is one of the issues closely tracked by central bank of a nation in implementing monetary policies. Essentially, central banks around the world often aim to control the inflationary pressures in a country, to avoid hyperinflationary business or economic environment from happening, as that would be dangerous and unhealthy for the society and economy of any nation (Bodie, Kane, & Marcus, 2007). As inflation rates are tracked by central banks, inflation rates are likely to affect monetary policies to be implemented by central banks, and thus, in certain situation, it can be used to predict stock returns – especially when the time the economy is overheating or underperforming (Yamarone, 2007). One of the most famously tracked economic indicator concerning inflation rates is Consumer Price Index (CPI). When CPI is high, inflation rate is high and vice versa (Moss, 2007).
Many studies on inflation rates and its linkages to stock returns are available. However, even in the context of United States, the empirical evidences on inflation rates to stock returns are mixed. According to Gertler & Grinols (1982), the addition of inflation to the standard two factor model of security returns is found to be able to enhance the explanatory power of the regression significantly in United States (year 1970:1 – 1980:1). Then, Kaul & Seyhun (1990) documented evidences that negative relations exists between stock returns with expected and unexpected inflation. This indicates that the relative price variability on the stock market is negative (from year 1947-1985). Nonetheless, according to Hasbrouck (1984) however, the hypothesized relationships between expected economic activities and expected inflation do not appear to be significant in explaining the negative relationships between expected inflation and stock returns in United States (year 1953:5 – 1979:5). Then, a study by Pearce & Roley (1985) found that there are very Limited evidences of an impact from unexpected surprises in inflation to stock prices on the announcement date (within research period from 1977:9 – 1982:10). However, the anticipated components of economic announcements do not significantly affect daily stock price movements in United States. In a similar manner, Chen, Roll & Ross (1986) found that unanticipated inflation was statistically weakly priced in explaining expected stock returns, during period when inflation is highly volatile. Besides, they also discovered that changes in expected inflation found to be statistically weakly priced in explaining expected stock returns, during period when inflation is highly volatile in United States (from year 1953-1983). McGowan & Dobson (1993) found that inflation rate is statistically and economically significant in explaining industry returns. Similarly, Abdullah & Hayworth (1993) found that stock returns are positively related to inflation (from 1980:4 – 1988:9). Besides, they also found that inflation Granger-cause stock returns and is able to explain a substantial proportion of the forecast error variance of stock returns.
However, apparently, the linkages of inflation rates to stock returns are no linear or simple. As found by Pearce (1984), the relationships of expected stock price changes and expected inflation seems to have changed over time (within 1954:12 – 1980:6). To explain, prior to 1972, investors apparently believed common stock to be a good hedge against inflation since they expected stock prices to rise at about the same rate as the general price level, holding expected real growth constant. However, in the period of high inflation since 1972, empirical evidences suggests that investors expected real stock returns to be adversely affected by inflation in United States. Later, Pearce & Roley (1988) demonstrated empirical evidences indicating that time varying firm characteristics related to inflation (i.e., debt-to-equity ratio and inventories, when FIFO inventory accounting is used) exerted dominance influence in determining the effect of unanticipated inflation on a particular stock returns (year 1977:11 – 1982:12). In their study, it is discovered that the effects could be either positive or negative in United States. From another perspective, Ferson & Harvey (1991a) found from their study that compensation for bearing inflation risk is highest near business cycle peaks (year 1964-1986).
In the international arena, empirical evidences on inflation rates to stock returns are mixed and inconsistent. Studies discovering statistically significant relationships between stock returns to inflation rates are as follow. For instance, Bodurtha, Cho & Senbet (1989) found evidences that the international unanticipated inflation is significant in explaining the cross-section of average stock returns in United States, Canada, United Kingdom, France, Germany, Australia, and Japan (year 1973:1 – 1983:12). In Israel however, Amihud (1996) found a negative and strongly significant relationship between unexpected inflation and stock prices by employing a market-based measure of unexpected inflation (1986:1 – 1991:10). Then, in Australia, Groenewold, O’Rourke & Thomas (1997) found negative relationships between (expected) inflation and stock returns in Australia (from year 1960:3 – 1991:9). They documented empirical proof that the negative relationships puzzle is found in the macroeconomic interactions: an increase in the expected inflation rate will increase equilibrium real output that has a negative impact on stock returns. In a similar way, Adragi, Chatrath & Raffiee (1999a) documented negative relationships between real stock returns and unexpected inflation, which persists even after purging inflation of the effects of the real economic activities in Korea and Mexico (from year 1978:1 – 1996:3). The researchers also found such relationships in Peru and Chile. Specifically, Adragi, Chatrath & Raffiee (1999b) also documented negative relationships between real stock returns and unexpected inflation, which persists even after purging inflation of the effects of the real economic activities in Peru and Chile (from year 1985:1 – 1995:12). In emerging country such as Nigeria, Udegbunam & Eriki (2001) also discovered empirical evidences strongly support the proposition that inflation exerts a significant negative impact on the behavior of stock prices (from year 1980-1997). Then, Patra & Poshakwale (2006) also found empirical results indicating that short run and long run equilibrium relationships exists between inflation and stock prices in Athens stock exchange (research period from 1990-1999). A comprehensive study on the relationships between inflation rates to stock returns also suggest that inflation rates are priced in stock returns in many countries around the world. For example, Patro, Wald & Wu (2002) found that country-specific macroeconomic variables such as inflation able to explain the beta (i.e., risk) of a country equity index returns (in 16 OECD countries). Then, similarly, Fifield, Power, Sinclair (2002) found that inflation is one of the economic factors able to explain returns in 13 emerging stock markets.
Contradictorily, there are also abundance of literature failed to identify any statistically significant relationships between stock returns and inflation rates. For example, Kwon, Shin & Bacon (1997) found that inflation related variables are not priced in stock return in Korean stock market, using various distributed lag regression, from 1980:1 – 1992:12. Then, according to Muradoglu, Taskin, & Bigan (2000), out of 19 emerging countries, there are only evidences showing inflation Granger-cause stock returns in Brazil. To explain in greater details, there are only evidences showing stock returns Granger-cause inflation in Jordan and Zimbabwe. Then, it is also found that there are only evidences of bidirectional Granger-causality between stock returns and inflation in Argentina. In a similar spirit, Davis & Kutan (2003) found that macroeconomic volatility, as measured by movements in inflation, has a weak predictive power for stock market volatility and returns (under research period from 1957:1 – 1999:4). The finding also suggests that there is no strong support for the existence of Fisher effect in the international context.
Overall, it can be seen that the relationships between inflation rates to stock returns are confusing, often contradictory, and can varies accordingly from time to time. Mixed empirical evidences are abundance. Perhaps findings from Ahmed & Cardinale (2005) can summarized the complexities of the relationships best. Specifically, they found that both the long-term and short-term aspect of the correlation between equity returns ad changes in consumer prices (in United States, Japan, United Kingdom & Germany). They found mixed support on the hypothesis regarding stable long run equilibrium relationships between inflation and stock returns, while strong evidences that indicate asymmetric behavior during different inflationary regimes in the short-term.
Economists such as Yamarone (2007) and Tainer (2005) have been arguing that Employment Situation Report (in United States) is one of the market shaking economic indicators. This is logical, as employment rate can be used to gauge the relative health of the economy. High unemployment rate often signify unhealthy economy. In contrast, low unemployment rate suggest that economy is booming and healthy. Lower unemployment, theoretically speaking, will translate into higher consumer spending (as everyone has a job and thus, disposable income). As asserted by Yamarone (2007), household spending is the driver behind economy of United States, as the household spending approximately contribute to around 66% of the entire United States economy output. Scholars agreed on the importance of employment situation to stock returns as well. Specifically, Kim & Wu (1987) found that labor market variables are one of the very significant priced risk factor in explaining stock returns in United States (year 1973-1985). Then, Flannery & Protopapadakis (2002) had studied and found that Employment Report announcement is one of the six candidates for priced factors in United States (year 1980:1 – 1996:12). Again, Graham, Nikkinen & Sahlstrom (2003) found that Employment Report announcement is one of the five significant macroeconomic announcements (with total of 11 announcements under-investigation) which influence stock valuation (year 1995:1 – 2001:12). In fact, Employment Report exerts the greatest influence among these announcements.
There is no lacking of literature concerning employment situation to stock returns. For example, Gertler & Grinols (1982) found that the addition of unemployment to the standard two factor model of security returns able to increase the explanatory power of the regression significantly in United States (year 1970:1 – 1980:1). Then, Huang & Kracaw (1984) discovered existence of unidirectional Granger-causality from stock returns to changes unemployment rate in United States (year 1962:Q2 – 1978:Q4). Similarly, Park (1997) found that employment growth rate exhibit the strongest negative influence on stock returns among some of the popularly known macroeconomic variables (year 1956-1995). Apart from that, they also found that employment growth is also more negatively related with future corporate cash flows when compared to other economic variables. Nonetheless, it is worthy to mention that scholars have also discovered that the influence of unemployment related reports may differ towards tock returns in different stage of the business cycle. For example, Funke & Matsuda (2006) found evidences in United States that there exist asymmetric reactions of stock prices to macroeconomic news. To explain, in a boom (recession) time, bad (good) news on unemployment may be good news for stock prices (year 1997-2002).