Currency Effects Often We Model Global Sales Simply Floor Not Only the EUR/CHF Rate Has As a Rough Figure, Typically in USD

Home , Swiss franc

Currency Effects Often we model global sales simply floor not only the EUR/CHF rate has as a rough figure, typically in USD. changed, but also the whole dynam- Maybe we say that a drug performs ic of the Swiss Franc with respect to similar like one already on the mar- any other currency. The Swiss Franc ket, assuming peak sales in the range simply got stronger overall. Now the of – let’s say – USD 1 bn to USD 1.5 Swiss Franc can move freely, which bn. But we might just as well make a wasn’t the case before due to the more thorough forecast using a SNB’s interventions (as can be seen (weighted) price, but applying it to very clearly in figure 1). the global pool of patients. Unless we really make a country-by-country analysis, we often use one currency. But how do we include different currencies in the valuation? How do we consider if two currencies drift apart? This is a generally neglected aspect of valuation. Figure 2: USD/CHF (source: Bloomberg.com) Comparing the share prices of some Case Study “Swiss Franc” important Swiss-quoted pharma and Switzerland is an export economy biotech companies we observe that and mainly exports to its neighbours all companies lost in value. that have the Euro. Since the Euro has become weaker and weaker – 1 Table 1: Share performance after the SNB aban- EUR was initially more than CHF 1.60 doned the EUR/CHF floor – Pharma and Biotech.

– the Swiss National Bank (SNB) has Company Loss after 1 day Loss after 2 days artificially weakened the Swiss Franc Novartis (8.7%) (13.7%) Roche (8.6%) (13.0%) to maintain an exchange rate of at Actelion (13.7%) (19.4%) Cosmo (6.4%) (8.8%) least CH 1.20. On January 15, 2015 Basilea (6.9%) (11.2%) Santhera (4.5%) (11.3%) the SNB abandoned that exchange Newron (6.1%) (11.8%) rate floor and the EUR/CHF exchange Molecular Partners (4.9%) (3.5%) rate dropped to about parity, which Addex (10.1%) (17.7%) was an instantaneous loss of about All companies lost in value. But the 17%. value loss was quite different.

FX Leverage The fact that the share values changed less than the actual currency is quite remarkable. It is fair to expect that Swiss companies have relatively more expenses in CHF than they have revenues in CHF. The rev- Figure 1: EUR/CHF (source: Bloomberg.com) enues come from worldwide opera- Clearly, with abandonment of the FX tions and Switzerland is with only 8 (foreign currency exchange rate) mn inhabitants a small market.

www.avance.ch However, on the cost side the head- er and more certain, accentuating quarters and some R&D facilities the leverage. Unfortunately, this weigh more heavily. They have so to idea is not confirmed by the data. speak an FX leverage. If we assume a Molecular Partners, e.g., with an out- company with the revenue structure licensed project in phase 2 as the as in table 2 then the value differ- most advanced project at the time, is ence after an event like January 15 least affected even though it should for the Swiss Franc leads to an even be affected most according to the stronger value loss than the FX above reasoning. Depending on the change would suggest. currency of the license contracts at Table 2: Example Company Valuation. least revenues from milestones might not be subject to changes in Before Event Revenues Expenses rNPV of CHF Cash Flows CHF 1 bn CHF 10 bn currencies. rNPV of non CHF Cash Flows CHF 99 bn CHF 70 bn TOTAL CHF 100 bn CHF 80 bn FX market vs. Equity market After Event rNPV of CHF Cash Flows CHF 1 bn CHF 10 bn So, does the equity market know rNPV of non CHF Cash Flows CHF 82 bn CHF 58 bn TOTAL CHF 83 bn CHF 68 bn better than the FX-market itself? In the example of table 2 the value Given that Swiss interest rates are drops by 25% even though non CHF lower than European interest rates, currencies only fell by 17%. Clearly, the Euro should get even weaker in this is not we could observe in Janu- the future. But there is a fundamen- ary 2015. Nevertheless it is fair to tal difference between FX markets assume that the FX leverage is dif- and equity markets. FX markets are ferent for the companies. highly liquid and are the playground of arbitrageurs, using the slightest Table 3: Share performance after the SNB aban- differences between interest rates, doned the EUR/CHF floor – other industries. spot rates and forward rates to make some risk-free gains. The equity Company Loss after 1 day Loss after 2 days Nestle (6.2%) (12.6%) market relies much more on opinions Transocean (8.3%) (11.6%) and so-called under- or overvalua- Straumann (16.9%) (27.7%) Swiss Re (5.6%) (8.7%) tions might persist for a long time. It Credit Suisse (11.0%) (19.2%) is not possible to simply hedge these UBS (11.7%) (17.2%) valuation inconsistencies away as the Other industries reacted differently. fundamental value drivers – the fu- The banks seemed to suffer a lot, but ture cash flows, are not traded as- we cannot draw any clear conclu- sets. It seems that the equity market sions, the reactions were too differ- has also given its opinion on the fu- ent – unless that all companies lost ture development of the Swiss Franc, value. Straumann, a medtech com- to some extent. Since the share pric- pany, has a lot of manufacturing in es did not react as strongly as the Switzerland, so even more costs are theory suggests, the equity market in expensive CHF. Therefore the FX expects the EUR/CHF rate to re- leverage is even stronger. bounce. Interestingly, this is what happened to some degree. In August An R&D company is certainly more 2015 one Euro is worth CHF 1.08 geared than a commercialising com- again. pany. The R&D costs in CHF are earli-

www.avance.ch Liquidity taken care of in the discount rate. However, this still does not explain Typically, it is a diversifying market the very different reactions of the risk that would decrease the beta share prices in tables 1 and 3. Of and therefore lead to a lower value. course, some differences stem form This is somewhat doubtful and other developments in the same pe- would lead to the following situa- riod, just like any other day. But a tion: In a currency of a small country final explanation might also be the (like, e.g., Iceland, Denmark, Switzer- limited liquidity in smaller biotech land) the value of a company would stocks. The two really liquid stocks of be higher than elsewhere. On the Roche and Novartis reacted in a very other hand it would be the same similar way. But smaller companies company, no matter where it is lo- such as Newron, Molecular Partners, cated (disregarding tax aspects). or Addex are not monitored and traded in the same way. Fewer inves- Another method would be to value tors look at these shares and as long all cash flows of the same currency as nobody buys or sells these shares and then translate that value with the price does not move. Therefore the current spot rate to the home the reaction is not as precise as for currency. This method would require larger stocks. – most likely – different discount rates for different currencies. We FX in valuation have seen that this is not the method But how should we make use of cur- of choice. rencies in our own valuations? Many analysts have been taken by surprise, The most reasonable method seems as a changed FX rate was in some to assume a term structure for each models not even an input factor. As FX (i.e. EURCHF, USDCHF, GBPCHF, we have seen, there is a considerable etc) and to change every cash flow effect, and this effect depends on at CF(t) with FX(t) to the home curren- least two important points: cy. For practitioners it is sufficient to 1. FX leverage assume a (constant) long-term ex- 2. Future development of FX change rate. This is what most ana- rates lysts and investors have done, assuming that the Swiss Franc will weaken A complete valuation model should again somewhat. therefore include costs and expenses in their respective currency and for Conclusion each currency an exchange rate de- Valuation of equity is less scientific velopment is requires (i.e. for than derivative instruments. But with EURCHF we need an estimate for a sudden change of FX rates we al- 2016, 2017, 2018, etc.). This is proba- ready touch the topic of term struc- bly more than even the most sophis- tures, which could also be applied to ticated models use. Even in text- the discount rate (but nobody does books the FX issue is neglected (very so). We are quickly faced with the much like taxes for R&D companies problem to which degree of detail or success rates). Some say that FX is we want to value a company or a an additional risk that should be project. Assuming fix FX rates and fix

www.avance.ch discount rates is in most cases com- pletely sufficient, but on January 15, 2015 it was not for Swiss companies.