19 January 2018 Group 2
Robin Döttling Academic year 2017/2018
Bachelor’s thesis Economics and Finance Semester 1, Period 3
Natural hedging against currency risks of non-euro currencies for multinational companies report in euros
Yuan Lu 11109017
Abstract
As one of the cheapest methods to manage the foreign exchange risk, natural hedging is used widely in international companies with offshore subsidiaries in other countries. As the subsidiaries are operated in local currencies which are different from the international reporting currency of the company on a consolidated level, it is crucial to hedge out the currency. However, previous research seldom shows interest in this hedging method. In this paper, the extent to which the firm can benefit from the natural hedging strategy in the area of currency risk hedging is studied. Expectations are that both the FX exposure and the volatility of potential FX gains or losses will be lowered after the implement of natural hedging. The counterfactual analysis is done on eight non-euro currencies followed by a longitudinal analysis to test the hedging effectiveness over the time on three currencies including a currency which is extremely volatile in the period studied. In the end, the findings match the expectations in general, but the case in extreme environments worths noticing.
1. Introduction
As companies expanded the business around the world, risks are generated along with profits (Solnik, 1974). The transactions done in currencies other than the currency that the company is used in their consolidated financial report is exposed to the risk of revaluing by a foreign exchange (FX) rate that could be wildly fluctuating based on different FX policies between countries, macroeconomic factors and other market factors. According to Phillips (1995), the currency risk is usually higher in large companies. Although geographical diversification in different countries could be advantageous for multinational companies on a consolidated level when the exchange rates used are not closely related (Makar, DeBruin, & Huffman, 1999), the risks are only partly hedged on a broad sense because of the diversification, while the changes in FX rates are unlikely to be netted off entirely against each other.
To hedge against the FX risks, currency derivatives are used in many firms with growth opportunities and strict financial regulations (Géczy, Minton, and Schrand, 1997). However, as the inter bank offered rates become much lower or even negative in many countries as an impact from the quantitative easing policy to stimulate the economy
(Christensen & Rudebusch, 2012), the cost of using natural hedging might be lowered. When the foreign currency to be naturally hedged can be borrowed at a lower interest rate than domestic currency, the firm can benefit from both lower funding rates and the reduction of the FX exposure (detailed mechanism will be explained in section 2.1). Hence, the preference of borrowing in specific currencies to implement natural hedging compared to the relatively costly hedging derivatives might be changed. The effectiveness of natural hedging of FX risks also becomes interesting to be investigated, especially for the period after 2010 that witnesses the impact of unconventional policies in Europe. In practice, managers are also unsure about which currency to hedge on and how to hedge. Therefore, the mechanism and the way to test the effectiveness will also be provided in this research as a reference.
In this paper, the following research question will be answered: To what extent could natural hedging be beneficial to an international company reported in euros by reducing the FX exposure and the volatility of potential FX gains or losses for non-euro currencies on a consolidated level?
Longitudinal quantitative researches are done to answer the research question. In the second section, the theories on natural hedging will be explained in detail, including the terminology, general logic and reason to use natural hedging, also the mechanism of a real practice in company ABC. In the third part, the methodology of the quantitative research will be explained, followed by the research outlines. Then, the counterfactual analysis will be
done based on the results from the descriptive statistics. A longitudinal analysis will also be done to explain the results found in the counterfactual analysis in detail with real world disturbance. Finally, the research results will be summarized and suggestions for future studies will be given in the conclusion part.
2. Theoretical framework 2.1. Literature review
DeMarzo and Duffie (1995) claim that FX risks need to be managed according to the
financial policies in the firm. Also, financial hedging has appeared to be more frequent since the innovations of financial instruments after 1985. If the FX risks are not hedged, the balances in foreign currency will be fully exposed to floating FX rates, and the translation of the exposure from a foreign currency to the local currency that the company is reporting in might fluctuates substantially (Houston & Mueller, 1988). As a result, the equivalent value of the foreign assets might be decreased and the drop in the equivalent values between the months is captured as unrealized FX gains or losses in the statement of profit and loss (P&L). As the end product on the P&L, the net profit might be affected.
Most researches on currency hedging are done in the cases that foreign exchange derivatives are used including swaps and forwards. However, as another option to hedge against the currency risks except for the financial instruments (Houston and Mueller, 1988), natural hedging in the area of currency risk management is rarely discussed in the literature before. The mechanism of natural hedging on FX is to reduce the FX exposure of a currency by offsetting the asset balances with liabilities (Houston & Mueller, 1988). In an ideal situation, if the asset balances are fully netted off by the liabilities, the FX exposure will be eliminated. In essence, if the bank does not charge extra money on the borrowings in new currency accounts and the interest rates of different currencies are the same, natural hedging is costless. If the interest rate of a currency is higher than the currency that the company is currently funded in, the cost of hedging the currency naturally is positive and vice versa. Therefore, the cost of natural hedging are not necessarily positive. For instance, the London inter bank offered rate (LIBOR) is higher than the euro inter bank offered rate (EURIBOR) in 2017, suppose that the bank charges the same spread on borrowing in both currencies, the interest rate of borrowing in euros will be lower than borrowing in ponds. In this case, a British international company can benefit from not only a lower euro interest rate by borrowing in euros and repaying the existing fundings in ponds, but also a reduction on FX
exposure in euros (if ponds are used as the reporting currency on their financial reports on the consolidated level). By contrast, hedging with financial instruments has a positive economic cost (Houston and Mueller, 1988).
2.2. Natural hedging practice of company ABC
Company ABC is an international company with offshore subsidiaries in 17 countries. Different subsidiaries, also known as local legal entities, have different local (reporting) currencies depending on local reporting and auditing policies. But when the company needs to report to shareholders as a whole, euro (EUR) is used on a consolidated level.
To fund all the local entities for their operations, a funding entity is set up to support as an in-house bank. Most of the external borrowings are done by the funding entity from the banks. Until now, the company only has bank accounts in EUR, pound (GBP), Canadian dollar (CAD), Swedish krona (SEK), and Norwegian krone (NOK), so only these currencies can be borrowed from the banks.
When the company has an asset in currencies other than euro, it will be revalued to euro equivalence in monthly reports on a consolidated level. Since the FX rates are constantly changing, the euro equivalence might vary a lot when revaluing it every month. As a result, the net income might also be changed for this reason every month. The amount needs to be revalued is the scale of the FX exposure. Only the balances in revaluing accounts on balance sheets will be revalued, where revaluing accounts refer to the accounts that will be revalued based on the exchange rates on a monthly basis. One example of non-revaluing account type is equity account, since equity is not exposed to FX. Hence, there is no need to revalue it. The pattern and accounts of balance sheets and the P&L are the same for all the local entities.
The FX rates used for revaluations are not the same between the balances on the balance sheet and the P&L. P&L accounts are revalued by the MOR currency exchange rate, which is the exchange rate used on the MOR billing system worldwide. Balance sheet accounts are revalued by the GAP rate, which is the Applicable Currency Exchange Rate Used in Central Administrative Departments according to the general accounting procedure (GAP). Correct exchange rate methods are suggested by the GAP depending on which circumstances the currency conversion is under. The choice of exchange rate methods includes the OANDA rate generated by the currency conversion function, the U.S. treasury current exchange rate, the Wells Fargo spot rate, and the Bank of America exchange rate.
If natural hedging is done before the end of a month, the FX exposure can be reduced by creating a short position in the foreign currency in the funding entity. In other word, since an asset is owned by the company in a foreign currency, a liability needs to be created to offset the FX exposure. To create the liability, the funding entity needs to borrow in that currency and the cash that the funding entity gets will be used to repay the loan in euros. Therefore, the total borrowings of the funding entity on a consolidated level do not change.
The FX realized gains or losses will appear if an amount of money is paid to an external party in a currency that is different from the domestic reporting currency. It is essentially the difference between the conversion rate of the bank (spot rate) and the MOR rate that is used in the companies’ system for normal P&L consolidated revaluations, when the FX exposure needs to be revalued. The gains or losses will be actually translated to EUR and will not be revalued later on since the amount is no longer in foreign currencies.
Natural hedging cannot be done if the firm (or the funding entity of the firm) cannot or do not want to borrow in certain currencies. The constraint might be that higher spread will be charged by the bank on illiquid currencies to shift their risk of holding an illiquid asset to the firm. The spread captures the risk of holding the currency and it will be added on a
normal inter bank offered rate to form the overall interest rate on borrowings in that currency. The more illiquid a currency is, the riskier to hold it and the higher the spread will be charged. By the time that the exposure has been hedged and possible FX gains or losses have been reduced, the interest payment with the higher spread rate also creates a higher interest expense. If the firm has to pay a much expensive interest while the FX exposure does not seem to be high enough to create a higher FX loss than the interest expense, the company would not choose to hedge that currency with natural hedging. Romanian leu (RON) is such a currency that is not hedged in company ABC. Czech koruna (CZK), and Polish zloty (PLN) are currently not hedged by the same reason before, but as the business in the Czech Republic and Poland expands, these two currencies will be able to be borrowed directly from the bank shortly after. The other reason that a currency is not hedged could be that it is pegged to the international reporting currency of the firm. For example, Danish krone (DKK) is pegged to EUR with a deviation range of positive or negative 2.5% in 2017. If the interest rate of borrowing in DKK is higher than 5%, DKK will not be hedged. Therefore, the currency policy is also a significant factor in the decision-making process of hedging activities.
If the hedging activity has not been done before the end of a month, any balance sheet exposure (i.e. amounts on the balance sheet of the funding entity that are in foreign currencies) will be revalued by the GAP rate during the revaluation process by the end of every month.
The difference of the revaluations of all the revaluing balance sheet accounts will be recorded as an unrealized gain or loss in P&L. By doing natural hedging, the exposure is reduced or eliminated, and thus, the volatility of unrealized gains or losses will decrease.
For instance, if the subsidiary in the UK has an asset of five million pounds, the situation before or after GBP can be hedged might be different. For the hedged position, GBP can be borrowed by the funding entity, so five million pounds will be borrowed from the bank to create a liability position and the five million pounds cash that the funding entity gets will be used to repay the equivalent amount of EUR loans which is around 5.2 million (see figure 1 & figure 2). If GBP is not naturally hedged, it usually means that no loans can be borrowed in GBP from banks, thus, not actions can be taken to reduce the FX exposure. The GBP balances will be revalued to euro every month, and this might cause variances on the net profit from month to month (see figure 3).
Figure 1. Hedged position
Figure 2. Natural hedging process (success in eliminating the FX exposure) funding entity UK entity + £ 5m EUR bank loan - ~ € 5.2m GBP bank loan + £5 m
Figure 3. Failure to eliminate the FX exposure
Similarly, all other currencies follow the same step when it started to be hedged naturally by external borrowings. All the external bank loans mentioned above are created under the revolving credit facility (RCF), where firms can borrow in multi currencies from the bank with accordance to the RCF contracts and the credit rating of the firm. Firms can use
FX exposure of £ eliminated
Same £ amounts of asset and liability are hold by the company EUR bank loan - ~€ 5.2m, cash - ~€ 5.2m
(Asset + £ 5m; liability + £ 5m, -~€ 5.2m ) FX realized gain/loss
(Asset + £ 5m, + ~€ 5.2m; liability + £ 5m)
Convert cash £ 5m to cash ~€ 5.2m (£ 5m EUR equivalence) Difference between conversion rate of the bank (spot rate) and MOR rate
Cash + £ 5m
(Asset + £ 10m; liability + £ 5m) To eliminate £ 5m cash GBP bank loan + £ 5m
Balances until now: (Asset + £ 5m; liability + £ 5m) Asset + £ 5m
To eliminate the FX exposure, liability also needs to + £ 5m
FX exposure of £ not eliminated
Different £ amounts of asset and liability are hold by the company FX unrealized gain/loss
Balances until now: (Asset + £ 5m) £ 5m revalued to € every month
Difference of GAP rates between previous month and current month Asset + £ 5m
RCF as an alternative to commercial papers for short term fundings, especially for firms with a credit rating lower or equal to the medium investment grade (i.e. S&P BBB or Moody’s Baa) as they are not required to access the market of commercial papers (Duffy et al., 2005). Medium term loans can be transferred from short term loans, and long term loans are also available under the RCF. Except for the interest that needs to be paid to the bank, a
commitment fee will also be charged by the bank for the unused part of the facility (Duffy et al., 2005).
According to the general natural hedging principals, any company using natural hedging strategy follows the same logic. The difference could appear in the use of funding entities, but it will not cause differences in the result. As long as the foreign assets are held by the company, they are exposed to the FX risk. When natural hedging is implemented,
external borrowings in foreign currency will be used to hedge out the exposure of FX.
3. Methodology
The research method used is a quantitative analysis. The counterfactual analysis will be done on eight euro currencies separately between the fact that the currency is hedged or non-hedged and the counterfactual scenario. The four non-hedged currencies are GBP, CAD, SEK, and NOK, while the unhedged ones are CZK, DKK, PLN, and RON. The data will be taken from the annual reports and other internal files from the company ABC after the year 2010, including the FX exposure to calculate the counterfactual data and recorded monthly FX gains or losses on P&L of local subsidiaries with non-euro local currency that mentioned above from 2011. (The confidential numbers will be rounded to the nearest thousand according to the confidentiality rules in company ABC.)
First, the standard deviation of the FX gains or losses will be calculated for all the currencies with the data analysis function in Microsoft Excel. SPSS can also be used to get the descriptive data. Next, the counterfactual analysis will be done on all eight currencies to test how effective is natural hedging on each currency compared to unhedged scenarios. The potential FX gains or losses for each currency will be calculated in the counterfactual scenario, and descriptive statistics will be fetched to compare with the ones in the first step separately for each currency. As a measure of volatility, the standard deviations are used to quantify the effectiveness between the hedged and unhedged scenarios in the same period. In addition, a longitudinal analysis will be done on changes in the volatility before and after GBP, SEK, and NOK are hedged. The percentage change in the volatility will be
decomposed into the part caused by natural hedging and the part caused by other factors regarding the specific currency in the real world to demonstrate the disturbance of exchange rate movements on the effectiveness of natural hedging. CAD is not analyzed in this part because it is hedged naturally once the Canadian subsidiary has been set up.
The detailed logic behind the counterfactual analysis for both hedged and unhedged currencies are given below. Full coverage of the FX exposure is assumed in the
counterfactual scenario of the unhedged currency, and the unrealized FX gains or losses will be zero. As for the hedged currency, it is more difficult to mimic the counterfactual scenario as if it is unhedged. The FX exposure should be equal to the total amount of the underlying asset in a specific currency that is exposed to the FX risk, in other words, the balances in the revaluing accounts explained above. Potential FX gains or losses will be calculated by multiplying the exposure with the GAP rate back to the corresponding time period.
Compared to the unhedged scenario, the volatility of FX gains or losses on the P&L that would affect the net profit in the end is expected to be reduced by natural hedging, same as the scale of the FX exposure.
By definition, the mechanism of the natural hedging strategy is identical for all the international companies in the world. The use of funding entity is not a must in this mechanism, as long as foreign currencies can be borrowed according to local laws and regulations. Thus, company ABC is a valid representative of international companies using the natural hedging strategy. In this case, the construct validity is fulfilled.
The external validity of this study is also relatively strong. The result on all the currencies are analyzed from two perspectives. For the volatility change caused by natural hedging, it is generalizable for all mentioned currencies because of the economic logic behind. Although firms might not be able to apply the theory and the exposure still exists, the
identical trend on eight kinds of non-euro currencies illustrates a general amendment for the imperfect world. Furthermore, even though the bargaining power of different companies can be different and the interest rate that they get from banks might differ, but on the consolidated level, only the comparison of currencies within one firm matters. For the other part of the result regarding to the FX market changes on each specific currency, the conclusion can also be generalized as all the companies face the same unique global FX market.
4. Analysis
In this section, the analyses are done with the data from company ABC in both hedged and unhedged currencies based on the methodology illustrated in the previous section.
In general, the standard deviations all decreased substantially in the same time period for all eight non-euro currencies (see table 1). The finding is consistent with the expectation stated in part 3.
Table 1. Values of standard deviations under hedged and unhedged scenario for 8 non-euro currencies
Currency Unhedged scenario Hedged scenario % change
GBP 2,122,714 155,621 -92.67% CAD 452,316 8,717 -98.07% NOK 179,706 129,685 -27.83% SEK 252,100 18,248 -92.76% CZK 47,129 0 -100% PLN 61,269 0 -100% RON 3,852 0 -100% DKK 3,213 0 -100%
For GBP, CAD, NOK, and SEK, the counterfactual scenario is the unhedged situation. The time range used to compare both scenarios for GBP is from March 2015 to December 2017 since GBP is able to be borrowed under the RCF as a part of external loans from the bank. The time range of CAD is between August 2016 and December 2017, directly after the Canadian subsidiary is established in the middle of 2016. The standard deviation of NOK and SEK is calculated in the period from September 2016, when natural hedging was
implemented for NOK and SEK. The same time period is taken both in the real and the counterfactual scenario for a currency to ensure that the trend of FX rates stays the same for a valid comparison.
As of CZK, PLN, RON, and DKK, the counterfactual scenario is the hedged situation. Since the RCF loan drawdowns are supposed to match the FX exposure as closely as possible, the external borrowings from the bank are assumed to be the same amounts as the FX
exposure for each of the four currencies. Therefore, the unrealized FX gains or losses will be zero, no matter how the FX rates change. The volatility of zero gain or loss is for sure a zero. The time range applied here for all four currencies are from January 2011 to December 2017,
because they have not been hedged until the beginning of 2018, and the counterfactual scenarios can also be mimicked for the whole seven-year period. The decrease in standard deviation is 100% compared to the unhedged position, which is in consistent with the 100% decrease of the FX exposure in these currencies.
The reason that the standard deviations are lowered is that the scale of FX exposure is reduced in GBP, CAD, NOK, and SEK, and is eliminated for CZK, PLN, RON and DKK. By applying the natural hedging strategy, outstanding assets in those currencies are matched with similar size of liabilities for the each currency. Therefore, the FX exposure decreases. Given that the FX unrealized gains or loss are calculated as FX exposure times the change in GAP rates between the current month and the previous month, even if the GAP rate changes dramatically during the month, the absolute values of the potential gains or losses are still supposed to be lowered compared to the scenario as if it is not hedged.
Among the reductions of volatility of all the currencies, NOK is an outstanding one with a reduction of volatility for less than one third compared to other currencies. The reason could be that the scale of borrowings are not matched with the size of assets in NOK. The mismatch could be caused by a timing difference as the FX exposure for a month can only be finalized after the month-end accounting process, and the reduction of exposure can only be done in the next month. Managers might also decide to wait until the FX rates are more favorable to prevent a relatively large realized hit on the P&L. As explained in section 2.2, the potential FX gains or losses are unrealized and could fluctuate. However, if the hedging activity has been done, the gains or losses will be settled as a realized impact on the P&L. The third possible reason could be the borrowing constraint in a specific currency. When the external loan from the bank in one currency reaches the upper boundary of the facility (RCF), extra time and effort will be needed to create a new contract, and the price of natural hedging might not be the same since the spread that the bank would charge can be different during a new round of negotiation process on the new contract. Same explanations can be given for the question that why the FX exposure does not exactly become zero in the hedged scenarios for GBP, CAD, NOK, and SEK.
Overall, natural hedging is proved to be effective in decreasing the FX exposure and the volatility of potential FX gains or losses on all eight non-euro currencies that are studied above, both from the theoretical perspective and the practical perspective of the outcome through the counterfactual analysis.
However, if the analysis is not done over the same time period for hedged and
as obvious. To validate the comparison between the two different time periods, the change in standard deviation will be decomposed to the part caused by natural hedging and the part of exchange rate movements in the FX market. If the currencies are not pegged to each other, the movements in different currencies are usually different. Even though same market factors might affect more than one country at the same time, the extent to which each currency is affected can be different, thus, the standard deviations could also be different.
As is shown in table 2, the decrease in standard deviation caused by the natural hedging is not enough to cover the increase in GAP rates for GBP and NOK. Thus, the standard deviation after hedging does not appear to be lower than in the period before natural hedging is implemented. But if natural hedging has not taken place, the standard deviation will be higher for the latter period than it is now.
Table 2. Values of standard deviations before and after natural hedging is implemented for 3 non-euro currencies
Currency Before hedging After hedging Total % change % change caused by hedging GAP rate changes GBP 16,932 155,621 819.10% -92.67% 911.77% NOK 112,566 129,685 51.21% -27.83% 79.04% SEK 31,106 18,248 -41.33% -92.76% 51.43%
The total time range of all three currencies are between January 2011 and December 2017, but the point of time that natural hedging came into force is not the same for each currency.
For GBP, the first time period is from January 2011 to February 2015, since the first GBP drawdown of external loan happened in March 2015, and the unrealized FX gains or losses was affected by the decrease in the FX exposure. Nevertheless, the news and presage of the fact that the UK might leave the EU (‘Brexit’) spreads shortly after that (Dhingra et al., 2016), especially after the general election in May 2015 (Inglehart & Norris, 2016).
Therefore, the drastic change in the exchange rate for the second period makes sense. The 93% decrease of volatility from natural hedging is relatively small to cover the change in the other direction in an extremely volatile situation like this, where the word “extremely” here is defined as the extent of the volatility change that is more than 9 times more than the original size. Although the FX exposure is reduced largely, the GAP rate changes so much that the
result after the multiplication of the remaining part of the exposure still overwhelm the size of the unrealized FX gains or losses before.
For NOK, the separation point is before or after (and including) September 2016 as explained in the previous analysis. The need of NOK was first raised up by the increasing oil price in 2016. However, by the end of January 2017, the concern of the illiquidity of NOK and the overheated Norwegian housing market dragged down the need of NOK substantially, causing the unrealized FX gains or losses almost 80% more volatile than the period before. By naturally hedging NOK, around 28% percent of the changes are relieved.
For SEK, the time period settings are the same as NOK, since the starting point of natural hedging on NOK and SEK are the same. Although SEK is around 50% more volatile than the unhedged period due to the (FX) market factors, the total standard deviation is still more than 40% less volatile than the period before.
The potential FX gains or losses that have been analyzed above are all unrealized benefits or costs as explained in section 2.2. The realized gains or losses are out of the control of managers through natural hedging as companies are not allowed to speculate. The need of the firms is usually to reduce the exposure and prevent potential losses in case the FX rates go out of the control rather than to earn money on the movements in the FX market through gambling. Furthermore, the speculations are not allowed in many companies, according to the risk management policies and higher level financial regulations.
To conclude, the effectiveness of natural hedging is high and the volatility of the uncontrollable FX movements on P&L is reduced compared to the scenario if no hedging activities are taken.
5. Conclusion
In conclusion, the counterfactual analysis proves that natural hedging is beneficial for a multinational company reported in euros through the decrease in FX exposure and the volatility of potential FX gains or losses on non-euro currencies. The longitudinal analysis provides the breakdown of the effectiveness of natural hedging in reality with disturbance from the most influential factor of FX market movements. Also, the fair comparison of the volatility changes is demonstrated when the FX market is highly volatile. The decreases in the volatility purely caused by natural hedging are mostly between 90% to 100%, which can be regarded as highly effective. The reasons for one currency with the volatility decrease
around a quarter of the original size are also explained from three perspectives: the timing issue, decision making process of realized or unrealized hit and the borrowing constraints. Still, there are some limitations in this paper. Firstly, the disturbance in the analysis before and after the currencies are naturally hedged is simplified as the change in the demand on currencies in the FX market only. However, the market trend might not be the only factor that differs between the hedged and unhedged periods. Therefore, some other factors could also contribute to the change in the volatility of a currency over the time. Secondly, although natural hedging helps to relieve the volatility to some extent in the periods that specific currencies become extremely volatile, the standard deviation still appears to grow in some cases. Therefore, to a certain threshold, it might be more beneficial for a company to adopt foreign exchange derivatives or other advanced financial instruments.
Future researches can be carried out by investigating in more factors that would affect the volatility of potential FX gains or losses through a regression with more factors, and the influence of each factor could be studied. Moreover, a logistic regression can be done to test the threshold where the firm would like to choose FX derivatives rather than hedging
naturally combined with a cost – benefit analysis. The similar analysis can also be done in a larger data set to confirm the results given above. It would also be interesting to observe if the same facts can be found in other non-euro currencies for other multinational companies reporting in euros or other foreign currencies verses the international reporting currency of a firm on the consolidated level.
Bibliography
Brown, G. W. (2001). Managing foreign exchange risk with derivatives. Journal of Financial Economics, 60(2), 401-448.
Christensen, J. H., & Rudebusch, G. D. (2012). The response of interest rates to US and UK quantitative easing. The Economic Journal, 122(564).
DeMarzo, P. M., & Duffie, D. (1995). Corporate incentives for hedging and hedge accounting. The review of financial studies, 8(3), 743-771.
Dhingra, S., Ottaviano, G., Sampson, T., & Van Reenen, J. (2016). The impact of Brexit on foreign investment in the UK. BREXIT 2016, 24.
Duffy, T., Hatzakis, M., Hsu, W., Labe, R., Liao, B., Luo, X., ... & Yang, L. (2005). Merrill Lynch improves liquidity risk management for revolving credit lines. Interfaces, 35(5),
353-369.
Géczy, C., Minton, B. A., & Schrand, C. (1997). Why firms use currency derivatives. The Journal of Finance, 52(4), 1323-1354.
Houston, C. O., & Mueller, G. G. (1988). Foreign Exchange Rate Hedging And SFAS No. 52-Relatives Or. Accounting Horizons, 2(4), 50.
Inglehart, R., & Norris, P. (2016). Trump, Brexit, and the rise of populism: Economic have-nots and cultural backlash (HKS Faculty Research Working Paper Series RWP16-026). Retrieved from https://faculty.uml.edu/sgallagher/Trump_Populism_Norris.pdf Makar, S. D., DeBruin, J., & Huffman, S. P. (1999). The management of foreign currency
risk: derivatives use and the natural hedge of geographic diversification. Accounting and Business Research, 29(3), 229-237.
Phillips, A. L. (1995). 1995 Derivatives practices and instruments survey. Financial Management, 24(2), 115-125.
Solnik, B. H. (1995). Why not diversify internationally rather than domestically?. Financial analysts journal, 51(1), 89-94.
Appendix
Descriptive statistics table - Hedged currency
+ UK + Norway + Sweden + Canada
Mean 1517 Mean -44823 Mean 5300 Mean 902
Standard Error 26689 Standard Error 32421 Standard Error 4562 Standard Error 2114
Median 0 Median -252 Median 0 Median 0
Mode 0 Mode 0 Mode 0 Mode 0
Standard Deviation 155621 Standard Deviation 129685 Standard Deviation 18248 Standard Deviation 8717 Sample Variance 24217763 391 Sample Variance 16818282 608 Sample Variance 332994 529 Sample Variance 759915 04
Kurtosis 14 Kurtosis 5 Kurtosis 4 Kurtosis 12
Skewness -1 Skewness -2 Skewness 2 Skewness 3
Range 1190609 Range 404751 Range 59361 Range 46231
Minimum -677577 Minimum -376171 Minimum -8601 Minimum -14291 Maximu m 513032 Maximu m 28579 Maximu m 50761 Maximu m 31940
Sum 51566 Sum -717167 Sum 84794 Sum 15342
Count 34 Count 16 Count 16 Count 17
Confiden ce Level(95. 0%) 54299 Confiden ce Level(95. 0%) 69104 Confiden ce Level(95. 0%) 9724 Confiden ce Level(95. 0%) 4482
Descriptive statistics table - Hedged currency (counterfactual scenario)
+ UK + Norway + Sweden + Canada
Mean -508537 Mean -37810 Mean -90246 Mean -67676
Standard Error 364042 Standard Error 44927 Standard Error 63025 Standard Error 113079
Median -1 Median -1 Median -1 Median -8
Mode #N/A Mode #N/A Mode #N/A Mode #N/A
Standard Deviatio n 2122714 Standard Deviatio n 179706 Standard Deviatio n 252100 Standard Deviatio n 452316 Sample Variance 45059156 68885 Sample Variance 3229426 2352 Sample Variance 6355430 4169 Sample Variance 20458970 9729
Kurtosis 5 Kurtosis 2 Kurtosis 5 Kurtosis 3
Skewnes s -1 Skewnes s -1 Skewnes s -2 Skewnes s 0
Range 12651751 Range 724783 Range 1031124 Range 2129290 Minimu -7780626 Minimu -458601 Minimu -850095 Minimu -1083755
m m m m Maximu m 4871125 Maximu m 266182 Maximu m 181029 Maximu m 1045535
Sum -17290259 Sum -604960 Sum
-1443942 Sum -1082816
Count 34 Count 16 Count 16 Count 16
Confiden ce Level(95 .0%) 740650 Confiden ce Level(95 .0%) 95759 Confiden ce Level(95 .0%) 134334 Confiden ce Level(95 .0%) 241022
Descriptive statistics table - Unhedged currency
+ Czech Republic + Poland + Romania + Denmar k
Mean 574 Mean 2820 Mean 270 Mean -671
Standard Error 5142 Standard Error 6685 Standard Error 420 Standard Error 351
Median 0 Median 0 Median 0 Median 0
Mode 0 Mode 0 Mode 0 Mode 0
Standard Deviatio n 47129 Standard Deviatio n 61269 Standard Deviatio n 3852 Standard Deviatio n 3213 Sample Variance 22211720 88 Sample Variance 37539235 84 Sample Variance 1483518 4 Sample Variance 1032564 6
Kurtosis 21 Kurtosis 26 Kurtosis 7 Kurtosis 23
Skewnes s 0 Skewnes s -1 Skewnes s 1 Skewnes s -4
Range 500516 Range 694709 Range 27598 Range 24682
Minimu m -259366 Minimu m -377082 Minimu m -10986 Minimu m -18645 Maximu m 241151 Maximu m 317627 Maximu m 16612 Maximu m 6037
Sum 48232 Sum 236855 Sum 22674 Sum -56346
Count 84 Count 84 Count 84 Count 84
Descriptive statistics table - Unhedged currency (counterfactual scenario) + Czech Republic + Poland + Romania + Denmark
Mean 0 Mean 0 Mean 0 Mean 0
Standard Error 0 Standard Error 0 Standard Error 0 Standard Error 0
Mode 0 Mode 0 Mode 0 Mode 0 Standard Deviatio n 0 Standard Deviatio n 0 Standard Deviatio n 0 Standard Deviatio n 0 Sample Variance 0 Sample Variance 0 Sample Variance 0 Sample Variance 0
Kurtosis #DIV/0! Kurtosis #DIV/0! Kurtosis #DIV/0! Kurtosis #DIV/0! Skewnes s #DIV/0! Skewnes s #DIV/0! Skewnes s #DIV/0! Skewnes s #DIV/0!
Range 0 Range 0 Range 0 Range 0
Minimu m 0 Minimu m 0 Minimu m 0 Minimu m 0 Maximu m 0 Maximu m 0 Maximu m 0 Maximu m 0
Sum 0 Sum 0 Sum 0 Sum 0
Count 84 Count 84 Count 84 Count 84
Descriptive statistics table - Hedged currency (before hedging) +
UK
+
Sweden + Norway
Mean 3007 Mean -2960 Mean -14950
Standard Error 2395 Standard Error 3772 Standard Error 13651
Median 0 Median 0 Median 0
Mode #N/A Mode 0 Mode 0
Standard Deviation 16932 Standard Deviation 31106 Standard Deviation 112566 Sample Variance 286689669 Sample Variance 967557628 Sample Variance 12671081486
Kurtosis 10 Kurtosis 4 Kurtosis 24
Skewness 3 Skewness -1 Skewness -4
Range 100095 Range 201694 Range 935351
Minimum -19694 Minimum -92423 Minimum -683724
Maximum 80401 Maximum 109271 Maximum 251627
Sum 150363 Sum -201260 Sum -1016592
Count 50 Count 68 Count 68
Confidence Level(95.0%) 4812 Confidence Level(95.0%) 7529 Confidence Level(95.0%) 27247
