International Parity Relations and Economic Shock : Evidence from Swiss Franc Unpegging

In this paper, we examine the association between the macroeconomic variables interest and inflation rate, and the expected spot rate of Swiss franc against Euro around the abandonment of the ceiling on the Swiss franc by the Swiss National Bank (SNB), January 15, 2015, using the two international parity relationships, International Fisher effect (IFE) and Purchase Power Parity (PPP). We use the regression analysis to examine the significance of change in daily interest rates and monthly inflation rates on the change in actual daily spot rates. While our empirical results indicate no significant statistical relationship between the two inputs before unpeg period, our additional analysis examining both the magnitude and directional deviation of the actual spot rate compared to the spot rate for both unpeg and peg period predicted using the two parity relations suggest that spot rates predicted based on the IFE relations, in general, is relatively better in predicting the daily spot rate compared to that of the PPP relations. Although both parity relations are not perfect models for predicting daily exchange rates, they can be used as a framework to guide financiers to gain additional insight regarding the short-term directional trend of the spot rate.


Introduction
Since the introduction of the peg, the Swiss National Bank (SNB) sold a total of approximately 500 billion Swiss franc to buy foreign currencies, which now sits on its balance sheet. The decision by the SNB was largely successful in stabilizing the franc against the euro and in turn Switzerland's exporting sector. The Swiss franc has been a very popular place to park cash. Amid the euro zone debt crisis, demand for the Swiss franc surged, pushing its value up sharply (Phillips, 2015). Recently, with the United States economy recovering strongly post financial crisis, the dollar has strengthened. In addition, with the impending crisis in Greece and the Eurozone slipping into deflation owing to lack of demand euro witnessed a sharp depreciation. Along with the US, UK and Japanese economy having successfully implemented the extensive bond buying program, the European Central Bank (ECB) announced the launch of a large scale government bond purchase program, Quantitative Easing, which could possibly increase the supply of euros in the economy and thus lead to a further depreciation in euros (See Blackstone and Maclucas (January 20, 2015 Wall Street Journal) from http://www.wsj.com/articles/switzerland-scraps-currency-cap-1421320531?mg=id-wsj). The economic conditions in both Eurozone and US put financial pressure on the SNB to act in order to maintain its currency and economy's competitiveness. Ultimately, on January 15, 2015, the central bank based in Zurich, SNB, announced the removal of the peg.
The move by SNB to unpeg the Swiss franc created a financial turmoil in the global economy and the franc shot up by 40% within minutes of removing the peg and finally settled at around 20% appreciation against Euros. As a result, hedge funds and investors incurred huge losses on their positions. This unexpected move by SNB poses the question as to where is the franc headed in the short term and would SNB be able to achieve its objective of taming the appreciation in the Franc relative to the Euro. The objective to peg was to prevent the franc from appreciating and protecting the country's exports. Now with the peg removed, it is of great concern for all market participants, investors and financial managers, as to how the Swiss franc will move in the short-term.
The main objective of this study is to examine whether the two international parity relations, international Fisher effect (IFE) and purchasing power parity (PPP), are viable frameworks that can be used to predict the daily exchange rates in the short-term. Surprisingly, there is a dearth of academic literature on the parity relations' predictability on the short-term spot rates, although decades of exchange rate studies have discovered many well-known empirical puzzles, in essence failing to connect floating exchange rates to their theoretical macroeconomic determinants, or fundamentals (Aliber, 1973;Chen and Tsang, 2013). Frenkel and Rose (1995) offer a comprehensive summary of the various difficulties confronting the empirical exchange rate literature. Sarno (2005) and Rogoff and Stavrakeva (2008) present more recent surveys. Because the measurement of market expectations is quite difficult, if not impossible, additional assumptions, such as a linear driving process for the fundamentals, are typically imposed in order to relate the exchange rate to its currently observable fundamentals (Mark, 1995;Engle and West, 2005;Chen and Tsang, 2013 among many others). While Hakkio (1992) argues that the short-term different political and economic shocks could drive the movement of exchange rates, we further extend the idea and maintain that short-term exchange rates can be looked by using the parity relations to understand the probable directional changes of exchange rates influenced by the political and economic shock that impact the movement of the exchange rate of interest. In addition, we scrutinize the implications of the actions taken by the SNB and ECB. The competing initiatives of the SNB and ECB and the actions they take will have profound short-term effects on key macroeconomic variables such as interest and inflation rate, and, in turn, the valuation of the franc. Under the IFE and PPP frameworks, we provide additional insight as to how changes in macroeconomic variables account for short-term movement of the Swiss franc relative to the euro.
The Swiss National Bank and the European Central Bank (ECB) have a mandate to control price levels using monetary policy. Their decisions impact the country's prevailing interest and inflation rates, which in turn impact the movement of the Swiss franc and Euro. Various market participants like hedge funds, multi-national companies, central banks and regulators who closely monitor the movement of these key macroeconomic variables are directly impacted by the fluctuation of the these currencies. Therefore, the ability to predict the short-term movement of a currency with maximum certainty is indispensable for all financial market participants.
Our initial results indicate weak (no) significant statistical relationship between the interest (inflation) rate, and the expected spot rate of Swiss franc around the Swiss franc unpegging. Our additional analysis examining both the magnitude and directional deviation of the actual spot rate compared to the spot rate predicted using the two parity relations, however, suggest that IFE relations performs better in predicting the daily spot rate than the PPP relations. We also find that although both parity relations are not perfect models for predicting daily exchange rates, they can be used as a framework to guide financiers to gain additional insight regarding the short-term trend of the spot rate. Our results further provide additional evidence supporting the idea that in the short-term different political and economic shocks will drive the movement of exchange rate, as suggested by Hakkio (1992) and others.

Swiss Economy, Swiss Franc Pegging, and Unpegging
The Swiss Economy is one of the most affluent nations across the world with a well-defined manufacturing industry and a highly developed service sector. Over the years, the country has been able to specialize in a highly skilled labor force at a low unemployment rate of around 3.4% (Dec 14, 2014). According to the world economic forum -Global Competitiveness Report 2014-2015, Switzerland ranks 1 st for the sixth consecutive year owing to its business sophistication, highly educated talent pool, innovation, efficient labor market and world class infrastructure, followed by Singapore and United States (http://reports.weforum.org/global-competitiveness-report-2014-2015/report-highlights/#map). The economy has grown at an average annual rate of 1.31% from 2008-2014 (http://www.statista.com/statistics/263611/gross-domestic-product-gdp-growth-in-switzerland/) (see Figure 1).
During the period 2008-2014 there was inconsistent economic growth largely due to the European sovereign debt crisis beginning 2008, which led investors to park their assets in Switzerland, the safe haven. Due to the increasing demand for the Swiss currency, the franc soared 35% against the euro from 2009 to September 2011 (http://www.futuresmag.com/2011/12/31/exploiting-franc-peg). The rapid appreciation of the Swiss franc became a serious concern for an economy heavily dependent on its exporting sector in industries such as chemical and medicinal products, machinery, instruments and watches (http://ec.europa.eu/trade/policy/countries-and-regions/countries/switzerland/). Ultimately, to prevent further long-term damage to its export and tourism industry, on September 06, 2011, Swiss National Bank (SNB) announced the decision to peg its currency at 1.20 Swiss Franc per Euro.
The franc overvaluation has been controlled since the introduction of the peg. To stabilize the franc relative to the euro, the SNB bought large amounts of euros, which is now sitting on its balance sheet. With the depreciating euro, SNB found it hard to retain its competitive advantage. The fact that SNB would continuously require Francs to buy Euros to maintain the peg exposed the Swiss economy to a huge exchange rate risk. Secondly, the stimulus plan by ECB was f even furth recovering currency m peg.

Literatu
Internation the state of competing but also sig rising inter and is on inflation ra in an econo growing an inflation is  (1995) that produced dismal results, especially in the short-term, Chen and Tsang (2013) conclude that the difference between two countries' yield curve can predict exchange rate movements and explain excess currency returns one month to two years ahead.
Having established the importance of these two macroeconomic variables, we examine the extent to which both interest rate and inflation impact currency exchange rates of a given country through the IFE and PPP relations can predict the short-term exchange rate movements. We note that our approach is consistent with previous research using the relative yield spreads to predict future spot exchange rate, such as Frenkel (1979Frenkel ( , 1981 and hypothesize the following.
Hypothesis 1: To the extent that the parity relations provide consistent directional guidance of future short-term spot rates, current spot prices can be used to predict the directional change of short-term daily exchange rate of the Swiss franc.

Data and Methodology
To analyze and better understand the short-term movement of the currency (CHF/EUR), the spot rate (CHF/EUR), Libor rate in Switzerland and Eurozone and inflation in Switzerland and Eurozone data was extracted using Bloomberg terminal and applied to the two international parity relationships -International Fisher effect and Purchasing Power Parity.
International Fisher effect (IFE) links the interest rates of home nation and foreign nation with their exchange rates so both the markets are in equilibrium and no arbitrage opportunities exist. Under this condition, the interest rate differential is equal to the forward and spot rate differential (interest rate parity) and the interest rate differential can be used to predict expected future spot rate (international Fisher effect). To the extent that forward rate is a unbiased predictor of the expected future spot rate, interest rate parity (IRP) can be equated to IFE to predict the expected future spot rate of E(S).
Where, i h= Nominal interest rate in home nation; i f= Nominal interest rate in foreign nation; F= Forward exchange rate (CHF/EUR); S= Spot exchange rate (CHF/EUR); and E(S) = Expected future, i.e., the next day, spot price. To predict the expected future spot rate for next day, we assume that forward is the unbiased predictor of expected future spot.
Purchasing Power Parity (PPP) suggests that the exchange rate between the currencies of two countries should be equal to the ratio of the countries' price levels. The exchange rate between two countries should be in such a way that an identical good in two different countries are priced has the same price when expressed in the same currency.
Where, I h= Inflation rate in home nation; I f= Inflation rate in foreign nation; and E(S) = Expected spot exchange rate (CHF/EUR). Based on the IFE relationship, the difference in interest rates between two countries will dictate the movement of the spot rate. Similarly for PPP relations, nations with lower inflation rates will tend to find their currencies appreciating against the currencies of nations with higher inflation rates.
To better understand the viability of the parity relations to predict short-term spot rates, we first examine, based on historical data, whether daily interest rate and monthly inflation rate differential between two countries really drive the movement of daily spot rates as the parity relations suggest using three different regression analyses as listed below, following Madura (2014).
We use the simple diff computatio I f )/(1+I f ), o admit that minimal m (see Table   In fact, du negative nu both Swiss also be for funds. For following F Where iN s rates are hi Thus we pe price from of Switzerl

Results
As stated monthly in Eurozone L are two key change in S on January Figure 3 sh based on th explaining -1. -1. -0. -0. -0.
-0.     The daily interest rate and monthly inflation rate were taken for Switzerland and Eurozone. These rates were taken as inputs to compute daily exchange rate predictions based on the idea the IFE and PPP relations are viable models to predict daily exchange rates. We computed daily exchange rates for the entire period using this method. Deflation occurs when the inflation rate falls below 0% (a negative inflation rate).   making Swiss franc value depreciating, helping Switzerland export business sectors. In the meantime, European Central bank (ECB)'s announcement of quantitative easing (QE) on January 20, 2015 was long expected by investors in the foreign exchange market, and so while the effect of QE is inherently long term, its short-term effect on euro is relatively not that substantial.
Based only on observation, the predicted spot rates based on IFE and PPP relationships seem to move relatively in sync with the actual spot rates. To further investigate whether the movement in the daily spot rates is driven by the IFE and PPP relationships, the daily interest rate and monthly inflation rate differentials were regressed against the percent changes in actual daily spot rates. For all periods analyzed, Table 2 Panels A and B show, in the short-term, there is no statistically significant evidence to conclude that interest rate and inflation rate differential drive daily changes in spot rates during before and after unpeg period, respectively, while for the entire year combining before and after unpeg period, the coefficient on interest rate differential is significant at five percent level (Panel A) and the coefficient on inflation rate differential is not significant (Panel B). Overall R 2 is also pretty low. Panel C shows that for entire period combining before and after unpeg period, the coefficients on interest rate differential is significant at the one percent level, and the coefficient on inflation differential is significant at 10 percent level. Before unpeg period, however, only the coefficient on inflation rate differential is marginally significant at ten percent level and for after unpeg period, due to small observation (only one month) problem, regression does not provide the coefficients. For the entire year, the interest rate differential is more statistically significant as a predictor of spot rates relative to inflation rate differential. Combined together, because the results in Panels A and C show statistical significance between the interest-rate differentials and the daily future spot rate, from a statistical analysis, the parity relations based on IFE can predict daily spot prices better than PPP when we combine before and after unpeg macroeconomic landscape.
Our regression results reported in Table 2 are generally supportive of the earlier findings of Kravis and Lipsey (1978) and Frenkel (1981) who find that PPP did poorly in explaining the behavior of exchange rates. Similarly, Adler and Lehman (1983) found that deviations from PPP follow a random walk, without exhibiting any tendency to revert to PPP. Abuaf and Jorion (1990) suggest that while deviations from PPP tend to be substantial in the short run, deviations from PPP appear to take about three years to be reduced in half. Abuaf and Jorion's (1990) findings, however, do not apply to our study as our focus is fairly short run, i.e., daily spot rate movement. In addition, our regression results based on IFE is consistent with previous empirical studies that while the ability of interest rate differentials is likely to support the tendency for these differentials to offset exchange rate changes (Giddy and Dufey, 1973;Aliber, 1973), the relationship between the interest rate differentials and the exchange rate differentials remains to be weak (Meese and Rogoff, 1988).
Hauner, Lee and Takizwa (2011), however, suggested that numerically (but not statistically) the PPP effect is strongest under a peg regime. Taking this idea, shifting the analysis to one that is more numerical, the relationship between the predicted spot rates based on the parity relationships and the actual spot rates was analyzed looking at the daily absolute value deviation of the two rates, predicted versus actual. Table 3 are statistical results from the analysis. Based on these results, under the IFE relations, the predicted spot rate was on average closer to the actual spot rate before the SNB removed the Swiss currency peg against the euro. Under the PPP relations, the deviation was smaller during the period after the SNB removed the peg. Looking back at Figures 2 and 3, the results seen under the IFE relations can be attributed to the near-zero Libor rate with very little fluctuation up until the removal of the peg. The relationship between a country's currency and interest rates are interdependent on one another. During the currency peg, when the SNB actively participated in the market, Swiss franc was held, for the most part, constant against the euro, a global currency. Consequently, the central bank was able to hold interest rates constant for the duration of the peg. Market participants also had this information at their disposal, including the ECB, and conducted their daily activities based on this information. Now, with the removal of the peg and the currency now freely floating, there is once again information asymmetry, resulting in greater volatility of the Swiss franc. On the other hand, the results seen under the PPP relations are more difficult to interpret. Based on observation, the inflation rate differential of the two countries is very volatile and hard to explain solely based on the PPP relations. Mathematically, the drop in average deviation of actual spot rate against predicted rate can be attributed to the fact that it is difficult to measure inflation on a daily basis. Since the inflation rate is only reported and captured monthly by Bloomberg, it is not a true indicator of the true increase/decrease in the level of prices. As a result, the daily inflation rate does not capture the actual movement in the macroeconomic variable -interest rate, since prices are sticky. Hence we conclude that it would be inadequate to predict the daily change in the expected spot rate based on inflation rate differential. However, as we suggest before, current inflation rates both in Switzerland and Euro zone are extremely low close to zero (in fact both have negative inflation rate in deflation), potential bias will be minimal. Also post the unpegging, we observe data for only thirty days, hence it is not enough data to determine the true deviations. Analysis of the absolute deviation of actual spot rates from predicted spot rates based on the international parity relationships alone doesn't indicate the ability to predict future spot rates. Especially, considering currency movement in general rarely fluctuates as dramatically as it did when the SNB removed the currency peg against the euro. Therefore, in addition to the study above, another test was conducted to evaluate the extent to which the two parity relationships can predict the daily directional movements of the daily spot rates. A summary of the results can be seen in Table 4 Panels A and B, which show the number of days in which the directional movements of the actual daily spot rate were consistent with the predicted daily rate, respectively, before and after the removal of the currency peg. The results indicate that the IFE and PPP relations can be used to explain, roughly, on average, the directional movement of the daily spot rate half the time. These results are supportive of our hypothesis 1. From the two tests conducted, focusing on the absolute and directional deviation of the actual spot price from the predicted spot price, while the two parity relations are unable to predict exact daily movement of spot rates both statistically and numerically, both parity relations do provide directional changes reasonably well (see Figure 4 and Figure 5 Panels A and B).

Conclusion
Based on the statistical and numerical studies conducted, in the short-term, while the international parity relations cannot predict the next day expected future spot rate precisely (CHF/EUR), they can, at least, inform the users with the trend or general direction. That being said, practically, there is still relevance in being able to forecast the general direction in which the spot rate, Swiss franc relative to the euro, moves based on the changing interest and inflation rate differentials between Switzerland and Eurozone. For investors and financial managers who are more concerned with the exchange rate trend and directional change find benefits using the two parity relations.
Another objective of the study is to understand and predict the implications of the actions taken by the SNB and ECB. Interestingly, because there is some statistical and numerical evidence from supporting the idea that the parity relations can be used to predict daily short-term rates, the results gathered supports a conclusion made by Hakkio (1992), who suggest in the short-term different political and economic shocks will drive the movement of exchange rates. Focusing the Swiss unpegging event happened on January 15, 2015, we find the following three empirical regularities. First, and most importantly, our empirical results based on the two parity relations of International Fisher effect (IFE) and Purchasing Power parity (PPP), suggest that future daily spot rates predicted based on the IFE relations, in general, is relatively better in forecasting the daily spot rate compared to that of the PPP relations. This finding is generally supportive of Frenkel (1979) and Hakkio (1992) in that daily short-term rates can be used to forecast future daily spot rate changes, at least in directional change, based on international parity relations. Second, the analysis of the absolute deviation of actual spot rates from predicted spot rates based on IFE seems to be lower than those from PPP. Third, the percentage of the number of days the directional change is consistent with the predicted spot rate based on IFE is slightly higher than the same percentage based on PPP. Combined together, our results suggest that consistent with our hypothesis 1, current spot prices, especially based on IFE, can be used to predict the directional change of short-term daily exchange rate of the Swiss franc.
Moving forward, this paper adds credibility to the idea that short-term exchange rates can be looked at using the parity relations to understand the probable direction of exchange rates and the political and economic events that may impact that movement of the exchange rate of concern. In addition, our analysis suggests that while SNB's sudden decision to abandon the ceiling on the Swiss franc caught markets totally off-guard, its short term effect on Swiss franc depreciation seems to be relatively more effective than ECB's announcement of quantitative easing on euro value at least in the short term slightly over one month between January 15, 2015 and February 24, 2015 between Switzerland and euro zone. Clearly, however, this does not mean that ECB's decision is not effective. In fact, a weaker euro in comparison with relatively stronger US dollar boosts the competitiveness of eurozone exports in global trade, a way for the ECB to buoy the region's growth (Iosebashvili and Zeng, 2015).