IIFET 216 Scotland Conference Proceedngs Effects of exchange rate fluctuatons on performance measures n the Norwegan seafood ndustry Workng paper Nyrud, Thomas; Bendksen, Bjørn Inge & Dreyer, Bent Nofma Norwegan nsttute of Food, Fsheres and Aquaculture Research Abstract: The seafood ndustry n Norway reles on sellng most of ts producton output n a global market. The Norwegan Krone (NOK) s a vulnerable currency subject to frequent fluctuatons, n recent years especally n connecton wth developments n the crude ol prce. The Norwegan seafood ndustry s compettveness s senstve to changes n the exchange rate towards mportant foregn markets. Large and sudden changes can therefore be challengng for both the ndustry and ts buyers. Exchange rate effects further create methodologcal dffcultes when attemptng to understand and analyze changes n frms proftablty, and product-prce changes. The paper analyzes how large exchange rate changes affect key performance measures n the Norwegan seafood ndustry. We develop a trade-weghted exchange rate ndex for the seafood sector, and fnd that the sector hstorcally has been sgnfcantly more vulnerable to exchange rate fluctuatons than the average for other export-orented ndustres n Norway. We further fnd that exporters of pelagc fsh have faced the hghest degree of fluctuatons, more so than both whtefsh exporters and salmon exporters. After establshng that the seafood ndustry s above average senstve to shftng exchange rates, we attempt to quantfy how sgnfcantly unstable currency valuaton can affect revenue streams. Specfcally, we take a closer look at the development n seafood export revenues through the perod from 212 up to 215. Through mathematcal decomposton, we derve the effects of three factors on ths development: shftng exchange rates, changes n export prces, and quantty changes. We fnd that yearly revenue from seafood exports grew by 22.4 bllon NOK between 212 and 215, and that we can attrbute approxmately 14 bllon NOK of ths ncrease to NOK deprecatng aganst major tradng currences such as euros and the US dollar. Introducton Norway s economy s small, open, and heavly dependent on the petroleum sector. Ths makes the country s economc development senstve to changes n global macroeconomc condtons, and especally the crude ol prce becomes an mportant factor. Snce the begnnng of the 197s, producton and export of ol and gas has been domnatng factors n the development of the Norwegan economy and socety. Exposure to petroleum markets and a general senstvty to external stmul makes the currency market value of Norwegan Krone (NOK) fluctuate wdely. Bernhardsen & Røsland (2) show that n the long run the value of NOK s hghly correlated wth the crude ol prce and the relatve prce level between Norway and other countres, whle n the short run t s also affected by nternatonal fnancal nstablty and the nterest rate dfferental towards other economes. Akram (2) further show how the value of NOK and the ol prce hstorcally has had a negatve, non-lnear, correlaton, where ts strength vares wth the level and trend n ol prces. Fgure 1 depcts the movements n the blateral exchange rate of US dollars aganst NOK, and n the crude ol prce. 1
USD/NOK (2Q1=1) USD/Barrel IIFET 216 Scotland Conference Proceedngs 18 17 16 15 14 13 12 11 1 9 8 2 21 22 23 24 25 26 27 28 29 21 211 212 213 214 215 216 14 12 1 8 6 4 2 Indexed USD/NOK exchange rate Europe Brent Spot Prce FOB (Dollars per Barrel) Fgure 1 Movements n the USD/NOK exchange rate and n the crude ol prce. January 2 - June 216 (Source: Norges Bank and U.S. Energy Informaton Admnstraton (EIA)) Exchange rates are often assocated wth the compettveness of a country s domestc ndustry. In general, a deprecaton of local currency mproves the compettveness of domestc export-based ndustry, as long as the ndustry s not heavly relant on mported nput factors. In case of an apprecaton the opposte holds true, decreasng the value of exported goods but at the same tme makng any mported nput factors relatvely cheaper. The Norwegan seafood ndustry exports approxmately 95 % of ts total producton, makng revenue flows vulnerable to exchange rate changes towards mportant foregn markets. At the same tme, producton s almost exclusvely based on domestc catch and aquaculture, reducng exchange rate rsk on the cost sde to a mnmum. One excepton s n aquaculture, where a substantal amount of fsh-feed s mported. Iversen et.al. (215) shows how costs related to fsh-feed has ncreased n recent years, and ponts at currency devaluaton as one explanatory factor. However, for the seafood sector overall, mport of nput factors s relatvely lmted n comparson to total producton value. The connecton between nternatonal compettveness and currency valuaton has clear support n economc theory. In practce, we also see several ndcators of ths connecton, albet not always as straghtforward as the theory suggests. In 22 for example, proftablty n the Norwegan seafood processng and exportng ndustry reached a hstorcal low. Bendksen (23) lnks ths to a strong apprecaton of NOK pared wth dffcult market condtons at the tme. Due to ths, many companes fled for bankruptcy through 22 and 23. In 28/29 when the fnancal crss struck, we saw proftablty droppng to a 2-year low n 28, whle rsng to a 2-year hgh n the followng year. The value of NOK fell markedly through the last half of 28, a fall drven by the crss and the followng recesson, whch decreased the prce of ol, as seen n Fgure 1. Ths sudden drop n the value of NOK was problematc for Norwegan exporters who had large sums pendng on long-term sales contracts. Many had sgned currency forwards that now came out wth a loss, and even though many of these contracts were yet to be realzed at the end of the fscal year, exporters stll had to record them n ther fnancal statements. Ths sgnfcantly worsened the sector s fnancal results n 28, at least on paper. However, throughout 29, NOK reganed some of ts market value and many of the losses from the prevous year were never realzed. Instead, exporters now had to reverse them n 29 s fnancal statements, leadng to a sgnfcant mprovement of that year s results. Recent years have seen a sgnfcant deprecaton of NOK aganst several major, globally traded currences. Fallng ol-prces and a declne n ol-related nvestments have been a drvng force 2
. 2. 3. 4. 5. 6. 7. 8. 9. 1. 11. 12. 13. 14. 15. 16 IIFET 216 Scotland Conference Proceedngs behnd ths development. The crude ol prce has been decreasng snce md-214, but costreducton campagns and the completon of several large projects facltated a declne n petroleum-sector nvestments, and n the value of NOK, even earler than ths. Contnued low prces has led to further cost-cuttng and capacty reductons n the sector, and to a further devaluaton of NOK. Enterng 213, the prce per 1 EUR was at 7.32 NOK, whle the prce per 1 USD was at 5.52 NOK. At the end of 215, the NOK had deprecated to 9.62 per euro and 8.81 per US dollar, a value reducton of 31 % and 6 % respectvely. Throughout ths same perod, the revenues from seafood exports, when measured n NOK, has ncreased substantally across most segments. Total yearly revenues from Norwegan seafood exports came to just over 52 bllon NOK n 212. In 215, they had rsen to almost 75 bllon NOK, an ncrease of 43 % n three years. There s some evdence that the NOK hstorcally has been overprced, and that the devaluaton throughout 213-215 served as an adjustment towards a more correct effectve exchange rate. Ths s n accordance wth the theory of purchasng power party (PPP) and the noton that blateral exchange rates over tme should move towards a level that equalzes the prce of an dentcal basket of goods and servces n any two countres. One such ndcator of currency msprcng s the Bg-Mac ndex, developed and annually updated by the London-based news magazne The Economst. The ndex clearly ndcates an overprcng of NOK aganst both USD and EUR throughout the last ffteen years, as seen n Fgure 2. Ths overprcng ranges between 6 % - 12 % aganst USD and 3 % - 8 % aganst EUR. What we further see s that ths dfference has decreased n recent years, as NOK has fallen n market value. 14 12 1 8 6 4 2 USD EUR Fgure 2 The Bg-Mac ndex. Prcng of NOK aganst USD and EUR. (Source: The Economst) We propose three research questons that ths paper wll explore further: (Q1) (Q2) (Q3) Norwegan export ndustry s senstve to exchange rate fluctuatons. How does the senstvty of seafood exporters compare to that of other export ndustry? Are they more or less exposed to ths type of rsk, or are there no sgnfcant dfferences? The seafood ndustry conssts of dfferent segments. Are there any sgnfcant dfferences across segments n senstvty to fluctuatng exchange rates? How strong are the frm-level effects of exchange rate fluctuatons? How are revenues and profts n the ndustry affected? 3
IIFET 216 Scotland Conference Proceedngs Frst, we calculate trade-weghted exchange rate ndces for the seafood sector and for three underlyng segments dvded by export speces. Whtefsh, Pelagc fsh, and Aquaculture are examned separately. We use these ndces to compare the seafood sector to other ndustry, and to compare across segments wthn the seafood sector. We then take a closer look at the devaluaton of NOK, usng the perod from 212 to 215 as a case, and attempt to estmate how these changes n the value of NOK have affected the revenue flows of Norwegan seafood exporters. Methodology We develop an exchange rate ndex for seafood as a geometrcally weghted average of trade currences used by Norwegan seafood exporters. Methodologcally we follow the dervaton of the trade-weghted exchange rate ndex (TWI) developed by the Norwegan Central Bank, whch covers the export revenues of the whole of Norwegan ndustry. A verson of Laspeyres ndex formula s our startng pont, expressed as follows: V t ( v t a N =1 v ), for all t =,, T (Eq.1) Where V t s the ndex value at tme t, v s the base level for exchange rate, v t s the level of exchange rate at tme t, and a s the weght of exchange rate n the base perod. We use the formula to calculate yearly ndces, whch we then chan together. The chaned ndex s the product of the ndces for each nterval: V t ( v 1 a N =1 v ) ( v 2 a 1 N N v 1 ) ( v T =1 =1 ) (Eq.2) v T 1 a T 1 In dervng the seafood ndex, we start by calculatng blateral exchange rate ndces for each of the relevant trade currences. We then splce these together usng the weghts a as shown n Eq.(1). The weghts represent each blateral ndex s share of the total ndex, calculated as export revenues n NOK from trade n each currency as a share of total export revenues n NOK. A weakness wth usng shares of export revenue n NOK to construct weghts s that these shares themselves are affected by changes n blateral exchange rates. For example would a deprecaton aganst US dollars frst cause a deprecaton of the total ndex through a drect effect. However, due to ths deprecaton the value of export revenues from dollar markets would ncrease when measured n NOK. Other factors held constant, ths would ncrease the sze of the weghtng for the deprecated NOK/USD exchange rate and thus further deprecate the total ndex. In calculatng the ndex weghts, we have only had avalable trade data from 24 up to and ncludng 213. For weghtng the exchange rate development n 2-23, we have therefore used trade data from 24. Smlarly, for 215 and the frst half of 216 we have used weghts based on trade data from 213. All other years follow the method portrayed n the formula wth monthly ndexed exchange rate development geometrcally averaged by whole-year, one-year lagged weghts. Ths follows the method used n calculatng the TWI, and, as we wll see n the next secton, allows for comparson. Usng the same method, we calculate sub-ndces for each of the three major sectors n the Norwegan seafood ndustry - Whtefsh, Aquaculture and Pelagc. Each wth sector-specfc currency weghts. As a mathematcal exercse, we can decompose seafood export revenues n NOK nto three components: volume, market prces n local currency, and the currency exchange rate. Ths s shown n equaton (3) (5): 4
IIFET 216 Scotland Conference Proceedngs y = f(v, p, e) = vpe (Eq.3) y + y = (v + v)(p + p)(e + e) (Eq.4) y = (vp e + v pe + vpe) + (v p e + vp e + v pe) + v p e (Eq.5) Equaton 3 descrbes export revenues measured n NOK as a functon of volume (v) multpled by prce measured n a gven tradng currency (p) and by the exchange rate between NOK and sad tradng currency (e). Equaton 4 shows how a change n y wll be a functon of changes n each underlyng varable, whle Equaton 5 s a verson of Equaton 4. The frst parenthess of Equaton 5 shows the solated effect of a change n each of the three varables, whle holdng the other two constant. The second parenthess shows the solated effect of changes n two of the varables wth the thrd held constant, whle the last part shows the solated effect of all three varables changng. In sum, these effects equal the total change n y. Through ths decomposton of the total change n revenue, we can now look specfcally at how much of the total change s due to changes n each varable, and how much s due to synergy effects when two or more of the varables change smultaneously. More smply put, we can use the equatons to analyze how export revenues change over tme, and derve how much of ther change s due to volume changng, to market prces changng, and to the blateral exchange rates changng. Results Table 1 shows the weghtng between tradng currences n aggregate Norwegan export ndustry, together wth weghts for the seafood ndustry and ts three bggest sub-sectors. Approxmately half of the settlements for seafood exports are n euro (49 %), whle one thrd s n US dollars (34 %). There are however large dfferences between the sectors. The Pelagc sector s hghly exposed to US dollars whle the sectors labelled Whtefsh and Aquaculture have a hgher exposure to the euro. There are clear dfferences between the weghtng of the seafood ndustry and of total ndustry. Whle seafood exporters receve approxmately 97 % of settlements n EUR, GBP, JPY or USD, these currences only make up 59 % of settlements n total ndustry. Table 1: Dstrbuton between tradng currences n total ndustry and n the seafood ndustry. Average for 24 to 213 for the seafood ndustry and average for 24 to 216 for all ndustry. EUR GBP JPY USD Other Total ndustry 4 % 1 % 3 % 6 % 41 % Seafood ndustry 49 % 6 % 7 % 34 % 3 % - Whtefsh sector 5 % 12 % 1 % 36 % 2 % - Aquaculture sector 57 % 5 % 8 % 27 % 3 % - Pelagc sector 19 % 1 % 14 % 65 % % Large overall exposure to US dollars and euro calls for a comparson of how each of them affect the total currency-portfolo. A statstcal comparson of varance n the two blateral exchange rates, based on monthly observatons from January 2 up to and ncludng June 216, reveals a statstcally sgnfcant hgher varance n the rate for US dollars. Table 2: F-test for equalty of varance n blateral exchange rates. NOK/EUR and NOK/USD, monthly rates January 2 June 216 N Mean Standard devaton F-stat p-value NOK/USD 198 6.8 1.16 5.69 <.1 NOK/EUR 198 8.13.49 5
IIFET 216 Scotland Conference Proceedngs 114 11 16 12 98 94 9 86 82 78 74 7 2 21 22 23 24 25 26 27 28 29 21 211 212 213 214 215 216 Seafood All ndustry Fgure 3: Trade-weghted exchange rate ndces, for seafood and for all ndustry, January 2 June 216 (Source: Norges Bank and own calculatons) The seafood ndustry has a much narrower currency-portfolo n trade than total ndustry does. Dversfcaton and not puttng all your eggs n one basket s common n fnance for mnmzng the volatlty and rsk of a gven nvestment-portfolo. Smlar to stocks and other fnancal assets, currency prces fluctuate accordng to market supply and market demand. The relatonshp between volatlty and dversfcaton therefore holds also when lookng at currency prcng. Accordng to portfolo-theory, the smaller currency-portfolo of the seafood sector should make t more vulnerable to fluctuatng exchange rates than the case s for total ndustry. In addton, the seafood ndustry has a hgher exposure to the more volatle US dollar-rate than other ndustry does. In sum, ths mples that the seafood ndustry should be more vulnerable to fluctuatng exchange rates than other Norwegan ndustry. Followng the method for ndex-calculaton descrbed n the prevous secton, we construct an exchange rate ndex weghted after tradepatterns n all seafood exports. Fgure 3 depcts ths ndex, together wth the correspondng ndex for all of Norwegan ndustry, as calculated by the Norwegan Central Bank. To test for dfference n varance between the two ndces, we perform an F-test for equalty of varances. Results are presented n Table 3. Table 3: F-test for equalty of varance. Trade-weghted exchange rate ndces for seafood and for all ndustry, monthly observatons January 2 June 216 N Mean Standard devaton F-stat p-value Seafood ndustry 198 9.86 8.5 1.92 <.1 All export ndustry 198 94.55 5.81 Through statstcal testng on the perod from January 2 to June 216, we fnd a statstcally sgnfcant dfference n varance between the TWI for seafood and the TWI for total ndustry, where the seafood ndex has greater varance. Ths answers our frst research queston. The seafood ndustry has hstorcally experenced a larger degree of fluctuatons and uncertanty n foregn currency markets, than the case s for Norwegan ndustry as a whole. Ths has affected prcng volatlty towards foregn markets and changes n compettveness, as well as the volatlty n value of nventory, recevables, fnancal nstruments and debt. 6
IIFET 216 Scotland Conference Proceedngs Next, we turn to our second queston, and look at whether there are any sgnfcant dfferences across segments wthn the seafood ndustry. For ths purpose, we have calculated separate ndces for three broadly defned segments Whtefsh exports, Aquaculture speces exports and Pelagc speces exports. Through further statstcal testng, we fnd that the ndex for Pelagc exports has had the hghest varance n the selected perod. As an explanatory factor, we fnd that Pelagc exporters have been relatvely more exposed to US dollars as a tradng currency than exporters of both Whtefsh and Aquaculture speces. Consequently, exporters of Whtefsh and Aquaculture speces have hgher exposure to the relatvely less volatle euro. We do not fnd any sgnfcant dfferences n the volatlty of the Whtefsh- and Aquaculture ndces. Table 4: F-test for equalty of varance. Trade-weghted exchange rate ndces for seafood and for all ndustry, monthly observatons January 2 June 216 N Mean Standard devaton F-stat p-value Pelagc sector 198 84.86 11.87 2.11 <.1 Whtefsh sector 198 92.55 8.17 Pelagc sector 198 84.86 11.87 2.61 <.1 Aquaculture sector 198 92.35 7.35 Whtefsh sector 198 92.55 8.17 1.24 >.5 Aquaculture sector 198 92.35 7.35 12 116 112 18 14 1 96 92 88 84 8 76 72 68 64 6 2 21 22 23 24 25 26 27 28 29 21 211 212 213 214 215 216 Whtefsh Pelagc Aquaculture Seafood total Fgure 4: Trade-weghted exchange rate ndces, per sector, January 2 - June 216 Fnally, we turn to our thrd research queston and take a closer look at the effects exchange rate fluctuatons can have on revenue flows. Usng Eq.5 from the prevous secton, we have decomposed the change n seafood export value from 212 to 215 nto three underlyng components volume changes, market prce changes, and exchange rates changes. A result summary for total seafood exports s ncluded n Table 5. These results derve from trade statstcs, and assume two necessary smplfcatons. Frst, we assume that EUR, GBP, YEN and USD are the only four currences wth whch Norwegan seafood s traded. Ths s a reasonable approxmaton gven the results of Table 1. Secondly, we assume that these tradng currences dvde themselves between export markets as expressed n Table 5. 7
Bllon NOK IIFET 216 Scotland Conference Proceedngs Table 5: Change n whole-year revenues from seafood exports between 212 and 215. By market/currency and by component. Bllon NOK. Currency Total Exchange rate Volume Prce Synergy effects All markets 22,4 14,1 2,7 4, 1,6 EU EUR 17,1 5,5 5,7 3,4 2,6 Great Brtan GBP 2,7,8,4,8,6 Japan YEN 1,1 -,2,1 1,1 -, Other markets USD 1,5 7,9-3,5-1,3-1,6 Gven these smplfcatons, we fnd that from an ncrease n export revenues of 22.4 bllon NOK between the years 212 and 215, 14.1 bllon connects wth changes n the effectve exchange rate for NOK. Further, 2.7 bllon connects wth growng volumes, 4 bllon wth ncreased prces, and 1.6 bllon wth synergy effects from two or more of these varables changng smultaneously. Fgure 5 llustrates the effects of changng exchange rates. Note that Great Brtan s ncluded under EU, and Japan under Other n the fgure, whle shown separately n the table. We see the total ncrease of 22.4 bllon NOK to the left (blue), together wth an exchange rate adjusted ncrease of approxmately 8 bllon NOK (red), leavng the estmated exchange rate effect as the dfference between these values. We perform smlar calculatons for the EU-market and ex.eu/other markets. 24 22 2 18 16 14 12 1 8 6 4 2-2 -4-6 Total EU Other Actual change n revenue (NOK) Exchange rate adjusted change n revenue (NOK) Fgure 5: Change n whole-year export revenues from 212 to 215. Actual change (NOK) and change after exchange rate adjustments. Dscusson/Conclusons In the long run, the frst and last lnk n the value chan of an ndustry wll see the largest effects of prce changes, exchange rate drven or otherwse. Ths prce transmsson wll however move wth a certan tme lag. Ths leads to varyng long-term effects n dfferent sectons of the ndustry. Combned wth constantly fluctuatng exchange rates, ths makes t dffcult to dentfy who the prce change affects most strongly. Further, a Norwegan fsh processer buyng raw materals under a strong domestc currency and low frst-hand prces wll proft f the currency unexpectedly deprecates before they export the fnshed products. Lkewse, f the domestc currency were to apprecate between the purchase of raw materals and export of fnshed products, the company 8
IIFET 216 Scotland Conference Proceedngs would face a loss on margns. Companes wth a hgher nventory turnover, say fresh-fsh exporters, s naturally less susceptble to ths type of rsk. Conventonal seafood producton such as dred fsh and clp fsh are on the other hand good examples of hgher-rsk segments wth a lower rate of turnover. In sum, dervng who benefts or suffers from shftng exchange rates s a complcated exercse wth many varables and no straghtforward answer. Our study underlnes the mportance of fscal and monetary polcy as ndustral polcy measures for strengthenng the compettveness of domestc ndustry. The analyss further shows that common measures of performance such as revenue and proftablty contan nherent weaknesses when appled to the seafood sector. In understandng why companes perform as they do, t s also mportant to consder other, underlyng explanatory factors. The Norwegan seafood ndustry, located n a small, open economy and wth most of ts producton output exported to a complex and global market, s vulnerable to shftng currency markets. To understand key challenges n the ndustry, we also have to take nto consderaton the effects of ths uncertanty. Especally wth recent years` ncreased export revenues. The mpressve growth of the ndustry mght seem dazzlng, but our fndngs ndcate that the weakenng of NOK mght have played a sgnfcant role. Whle consderng ths, we also see that revenues n some segments and markets stll are strong even after adjustng for exchange rate effects. Wth the deprecaton of NOK n recent years, Norwegan ndustry has now passed a perod characterzed by a hstorcally strong domestc currency. A strong NOK through large parts of the 2s put the ndustry n a challengng global compettve poston. However, measures taken durng ths perod has put the seafood ndustry n a poston where they now, wth a more correct valuaton of NOK, can harvest the benefts from adjustments made. References Akram, Q. F. (2). When does the ol prce affect the Norwegan exchange rate?. Workng paper nr. 8. Norges Bank Bendksen, B. I. (23). Drftsundersøkelsen fskendustren Oppsummerng av nntjenng og lønnsomhet 22. Report 15/23. Nofma, Tromsø Bernhardsen, T. & Røsland, Ø. (2). Hvlke faktorer påvrker kronekursen?. Penger og Kredtt 3/ Dreyer, B. & Bendksen, B. I. (21). I etterpåklokskapens lys. Fnanskrsens effekter torskesektoren. Report 23/21. Nofma, Tromsø Iversen, A., Hermansen, Ø., Andreassen, O., Kongsvk Brandsvk, R., Marthnussen, A. & Nystøyl, R. (215). Kostnadsdrvere lakseoppdrett. Report 41/215. Nofma, Tromsø Norges Bank (2). Revson of the trade-weghted exchange rate ndex (TWI). www.norgesbank.no. Statstcs Norway. Onlne databases for seafood exports. www.ssb.no. 9