URN:NBN:fi:tsv-oa4445 Economic geographic characteristics in the Finnish paper industry − a case study ESA HÄMÄLÄINEN Hämäläinen, Esa (2011). Economic geographic characteristics in the Finnish pa- per industry − a case study. Fennia 189: 2, pp. 63–75. Helsinki. ISSN 0015- 0010. The main purpose of this paper is to reveal the Finnish paper industry from the economic geographic perspectives. There have been many mill and machine line closures especially in Finland after 2001. Therefore, it is interesting to re- search the development of manufacturing and transportation costs and paper prices during 2001−2008 in a large case mill. The research tradition of eco- nomic geography concerning the paper industry is scanty in the Nordic coun- tries, and not many discussions have been published. This paper attempts to narrow the gap between theoretical and empirical discussions concerning the paper industry. The empirical data is obtained from one large integrated mill, and the research data covers cost components from the years 2001–2008. The results show that the economic performance has lowered clearly in the case mill. An interesting finding was that in overseas distant deliveries, transit costs can even decrease due to inexpensive sea transportation and paper prices slight- ly increase, probably due to lowered competition in the exported paper quali- ties. Although the mill data has been examined in detail, it only covers one large paper mill with several machine lines. Therefore, the results can only be gener- alized to some extent to other export-dependent paper industries operating in peripheral areas with minimal local demand. Our study shows that the empiric methods of economic geography offer interesting views highlighting such spatial heterogeneities in the paper industry. The industry’s location affects competition through transit costs, and this topic should be included more in location and economic studies. Keywords: economic geography, paper industry, performance, transportation, costs Esa Hämäläinen, Department of geography and geology, FI-20014 University of Turku, Finland. E-mail: esa.hamalainen@utu.fi Introduction Since 2002, Finnish paper mills have been facing particularly challenging times concerning their ex- istence and operation. The paper companies have closed over 20 machine lines, including several paper mills (Oinonen 2008; Tappi 2008; Hetemäki & Hänninen 2009). During the past few years, the economic situation in the Finnish paper industry has been affected by lowering demand (Forestin- dustries 2009). The purpose of this empirical case study is to examine in detail the performance and the devel- opment of a Finnish paper mill through economic variables during the past years. It is described how a paper mill’s economic factors, like transport costs, behave in a spatial context and how these factors influence the mill’s gross margin. This sur- vey, from the viewpoint of the case mill, may help us to better understand the economical transfor- mation in the Finnish paper industry. The topic of this study, the paper industry, has been examined marginally in Finland from the perspectives of both economy and geography to- gether. Under these disciplines, researchers have published few papers in Nordic research journals (see e.g. Lähtinen 2007; Arlbjørn et al. 2008; Kos- kinen 2009). Hämäläinen and Tapaninen (2008) found that transportation costs increase in the spa- tial context in the European market. Later, 64 FENNIA 189: 2 (2011)Esa Hämäläinen Hämäläinen and Tapaninen (2010) showed that there are big monthly variations in paper deliver- ies at the country as well as the customer level. These variations in customer deliveries are proba- bly one reason for the long and costly warehous- ing times, which Koskinen and Hilmola (2008) and Koskinen (2009) presented in their papers. Basically, the Finnish paper industry was de- rived and developed because of many competitive advantages (see e.g. Diesen 1998; Dieter & Englert 2007): the availability of reasonable priced fiber; inexpensive hydro energy; large, modern and effi- cient paper machines; skilled personnel; and growing demand in the main markets until the late 20th century. The value of forest products was ap- proximately half of the total export in 1995, and in 2008 it was still around 18% (Forest industries 2009). In many rural areas of Finland, a local pa- per mill is the only significant employer in the community. Finland is a member of the EU mon- etary union, and Forslid et al. (2002) stress that a common worry in the Nordic region is that eco- nomic integration may lead to the loss of indus- tries and jobs in the peripheral regions (see also Krugman 1991). Therefore, it is interesting to study the paper industry from empirical and economic geographical perspectives. The subject is novel and it is particularly concerned with paper mills operating in peripheral areas with minimal local demand. The main research objectives are: • to study the characteristics of the home market and export market through delivered tons and profits; • to examine the development of economic per- formance in euro per machine hour; • to examine the development of paper prices, manufacturing costs and transit costs; and • to study the correlation between transportation costs vs. gross margin in an example European market. In this study, the main economic geographical research topic, transportation, is understood as a process in which the packed paper rolls are trans- mitted from the mill warehouse to the customers with common transportation means. The key eco- nomic variables are paper prices, variable and fixed costs and especially transportation costs. The transportation costs include all functions from the mill to the end consignees. The variable costs in- clude timber, fiber, chemicals, energy, and logis- tics costs during the manufacturing process (in- cluding packaging costs). All the costs and prices and their correlations are calculated at the country level to explore the interesting economic geo- graphic differences. Gross margin is the margin remaining when transportation, variable and fixed costs are subtracted from sales prices. The structure of the paper is as follows: Theories of economic geography, which contributed to the ideas in this study, are presented in section 2. In section 3, a closer look at the paper mill’s value- added functions is taken. The used case study methods, the origin of the data, the data mining and methods are described in section 4. The de- tailed results, which are based on empirical data analysis, are presented in section 5. Discussion and conclusions are summed up in section 6, and finally, in section 7, some future research topics are introduced concerning the paper industry in the Nordic countries. Theoretical background Economic geography as a research method In this section, economic geography is briefly re- viewed from two interesting and relatively novel angles: traditional economic geography and new economic geography (NEG). Both of these re- search perspectives have contributed to the ideas in this paper. Generally, the term “economic geog- raphy” refers to the study of economic activities from a geographical viewpoint (Martin 1999; Scott 2004, 2006). The discipline studies where eco- nomic industrial activities are located and why they are located in those areas. Economic geogra- phers generally use multidisciplinary methods, like mathematics and statistics, in order to better understand the processes found locally and glo- bally. Ever since Weber (1909/1929), there has been a permanent interest in the transportation aspects of the location problem of the firm; transportation fa- cilities are considered a major location factor, and the minimization of total transportation costs is re- garded as a basic objective. Isard (1956) has been concerned with the impact of concave transporta- tion costs on the firm location. In economic geog- raphy, there is a large number of relevant contribu- tions, such as the so-called German location re- searchers. These theorists provide general location theories, such as Weber’s transportation orienta- tion, market (and purchasing) area theory and Lösch’s spatial designs. FENNIA 189: 2 (2011) 65Economic geographic characteristics in the Finnish paper … Alfred Weber’s work (1909/1929) is considered to have grounded the roots of modern location theories. Particularly Weber assumed that firms will choose a location in order to minimize their costs. Weber’s model assumes perfect competi- tion, implying a high number of firms and custom- ers and a perfect knowledge of market conditions, both for the buyers and the suppliers. At present, it can be considered that the European paper market is obviously in a state of perfect competition in many qualities due to lowering demand and over- supply (RISI 2009). According to Weber, three main factors influence industrial location; trans- portation costs, labor costs and agglomeration economies and location thus imply an optimal consideration of these factors. Activities having a high level of use of raw materials tend to locate near supply sources, such as aluminum factories locating near energy sources (electricity) or port sites. Aiura and Sato (2009) show that firms try to differentiate geographically in order to avoid price competition and that firms have an interest in lo- cating in the proximity of the raw material site in order to reduce transportation costs. The existing paper industry in Finland and other Nordic countries was originally built near hydro energy locations and raw material sources. The material index (Weber 1909/1929) in a paper mill is approximately 1.5–3, as around three tons of timber and other materials are needed for a ton of paper. Later, the Finnish mills were placed near ex- port harbors due to logistics reasons. The 2008 Nobel Prize winner Krugman (1991, 1995) has stressed that geographic economic is- sues have been at the periphery of mainstream economics for a long time. Krugman considers that in a world characterized both by increasing returns and transportation costs, there will be an obvious incentive to centralize manufacturing near its largest market. The simple reason is that by locating the production near the largest market, one minimizes transportation costs. Location and economic geographical character- istics have been regarded as irrelevant factors until now in many theoretical economic studies (Arbia 2001). In a recent study, Behrens et al. (2009) re- mind that although transportation costs are a key ingredient of economic studies, the transportation sector is usually abstracted, and freight rates are taken in as parametric and are not set by the mar- ket. Bullock and Cliff (2004) stress that purely eco- nomic discussions say little about geographical space and its relation to the adaptive behavior of businesses. Essletzbichler and Rigby (2007) point out that the economic performance of regions fluctuates because of differences in their charac- teristics. Regions, economy and companies are not static entities, and this environment may be radically transformed, especially in foreign trade dependent firms counting on demand develop- ment and changes in consuming behaviors. This type of a process can be observed to be happening in the paper industry due to the news media head- ing in a digital direction and the traditional paper media suffering in the future (Hetemäki & Nilsson 2005). Kilkenny and Thisse (1999) point out that com- panies exporting their production outside the re- gion in which they are located must set a price that is high enough to cover trade costs but that is low enough when compared to local competitors. Ot- taviano et al. (2003) state that when trade costs are sufficiently low, firms are able to export profitably to a foreign region. The high transportation costs could be naturally surpassed with higher econom- ic efficiency than that of the competitors, which is often neglected in economic geographical studies. As a summary, it seems there is also a place for traditional geographical examinations that are based on empirical data. It can be noted that eco- nomic geographical research methods reveal es- pecially interesting space-linked issues. This em- pirical orientation also provides some ideas for discussions that put their main effort into explain- ing the complexity of space with purely theoretical assumptions and models. Empirically oriented ap- proaches might be better compromises for the pur- poses of studying spatially diffused objects, espe- cially those that have not been examined in detail before. Some characteristics of paper mill functions An integrated paper mill consists of manufacturing and a complicated mixture of transportation and cost management functions. Figure 1 presents the value chain factors, which in this paper are under- stood as covering the process functions from the timber yard to the end customers. Customers order specific machine-based production grades. These qualities have unique production tons per hour, required moisture, surface, gloss, base weight, density, color and predefined roll width for the printing machines. The actual costs are calculated 66 FENNIA 189: 2 (2011)Esa Hämäläinen by summing up all the manufacturing costs and transportation costs of the sales on a monthly ba- sis. The fixed costs are allocated to the grades by the production performance of the grades in tons per machine hour. The exporting of the paper products from Fin- land to the markets occurs in many stages: storing at the mill’s warehouse; short domestic truck or train transportation; domestic port transportation; intermediary converting and repackaging; inter- mediary warehousing; and the final local transpor- tation to the customers. These distribution centers are located abroad, where consignees pick up their rolls. The exported paper qualities are mainly so-called improved grades, which usually have a lower supply and therefore higher prices. During the 80s and 90s, Finnish paper compa- nies expanded by merging with e.g. European pa- per producers to acquire market share and to be more competitive closer to the market (Lamberg & Ojala 2006). The key idea was to precisely control the market, supply and prices. The paper mills be- came more like production units than profit cent- ers. Certain paper grades, e.g. office bulk papers, could be produced not only in one but in many mills and delivered to customers from the closest one. The Finnish paper machines could stay in competition by being efficient with large units. The export transportation of forest products is a top priority for today’s executives of the forest products industry (Hölsä 2005), but the whole supply chain has not been extensively studied us- ing a holistic approach. Koskinen and Hilmola (2008) believe that logistics does not support the hoped improvements in the paper industry, be- cause paper rolls remain in the inventories for ap- proximately 45 days. There is room for the paper industry to improve its logistics management in or- der to decrease capital costs during inventories (Gallis 1997). Lehtonen (1999) and Hameri and Lehtonen (2001) also present comparable conclu- sions in their study. According to them, speedier transit operation easily generates direct cost sav- ings amounting to 2–5% of annual turnover. Trans- portation problems, especially the long and costly warehousing and slow deliveries that Eloranta et al. (1994) observed in the early nineties, still exist in 2008 (Koskinen & Hilmola 2008). The Finnish bulk paper mills are usually inte- grated units that have both fiber and paper ma- chine lines. Due to the high productivity of the machines, especially the fixed costs in euro per paper ton are usually lower than those of the com- petitors. This has helped Finnish mills to overcome rising distance costs but not the currency chal- lenges, like the relation of the euro to the U.S. dol- lar and the Swedish krona. Empirical data and survey methodology Lewis (1998) considers that case studies offer a po- tentially efficient means for comparing complex Fig. 1. The origin of the re- search data: the case mill’s basic data sources, the mill database (allocations) and the reporting modules. FENNIA 189: 2 (2011) 67Economic geographic characteristics in the Finnish paper … and disparate operation settings. Their iterative tri- angulation employs systematic iterations between literary review, interviews, case evidence, and in- tuition. The case study methods are well suited for producing context-dependent knowledge (Flyv- bjerg 2006), and case study research can provide microeconomic discoveries which are not possi- ble to acquire through other methods (McCutch- eon & Meredith 1993). Ballou (2001) reminds that the acquisition of correct and reliable data is cru- cial to the successful practical application of any model. Even the best model, if implemented with incorrect data, will produce false supply/demand chain configurations. According to Baxter and Chua (1998), one of the greatest practical difficul- ties in conducting case study research is the lack of access to field sites. Yin (2003) emphasizes that the “distinctive need” in conducting case study re- search lies in the “desire to understand complex phenomena”. Economic geography with multidisciplinary methods seems to fit well for research purposes where the target is to study how an exporting in- dustrial site functions both spatially and economi- cally. The research ideas of this study are based on theories from microeconomic and transportation geography combined with statistical analysis. The target of the study is a single exporting mill located in Southern Finland, close to an exporting port. The mill manufactures all of its products locally in a certain region, but acts very globally because of the high export dependency and transportation of the final products to dozens of countries. This spa- tial connection in particular separates economic geographical studies from the purely economic ones. The empirical data covers exhaustively the pe- riod between 1/2001 and 12/2008 (96 months). We obtained the data from the case mill’s cost management SQL database (Fig. 2). This extensive quantitative database holds the mill’s past and present information, which is affirmed by eco- nomic controllers and official authorities. The fig- ures have been checked and reported in the mill’s annual economic reviews. The machine hours used in this study are net hours, when the machine has been in production. There are fixed and vari- able costs as well as spatial aspects, such as trans- portation costs and paper prices to points of desti- nation. The variable costs also cover transporta- tion costs of timber from forests to the timber yard at the mill. Calculated transportation costs of the final product cover total logistics costs from the mill warehouse to the customers. These figures have been transformed into a single large dataset based on packed net tons and euro. Within the da- taset, the processed figures are proportionate, val- id and fully intercomparable, and all data is linked to the same time period, 2001–2008. The param- eters are calculated mainly in €/t or in percentag- es, which gives the possibility to make relevant comparisons. The results of the calculations are shown as ratios to maintain business confidential- ity, and the key objective was to determine the rel- evant value-added factors. The research data cov- ers prices, delivered tons and transportation up to 73 countries, tens of different paper grades, and millions of delivered paper tons. From the view- point of triangulation, we carried out a number of interviews and discussions with the financial man- agement of the case mill and additionally with mill personnel in other local and foreign mills. These discussions significantly helped us focus our re- search on the relevant topics. The research data was transferred into Excel spreadsheets and into Stat 4.1 for statistical corre- lation analysis. The correlation was calculated to reveal the dependency between the transportation costs and the gross margins. All the economic fig- ures were reported out for research purposes only. Our study attempts to link the industrial and time- based performance values with geographical di- mensions, which is not a very common research Fig. 2. The paper mill’s demand-supply chain. 68 FENNIA 189: 2 (2011)Esa Hämäläinen subject, when considering the Finnish paper in- dustry. Findings and results In this section, the research data of the case mill is examined to find some typical and specific eco- nomic characteristics of the Finnish paper industry from different perspectives. Firstly, a closer look is taken from the market’s point of view, and figure 3 presents the monthly development of both the share of the home market of the total mill sales and the share of the profits during 2001–2008. The case mill is a typical Finnish mill exporting around 90% of its total production. Figure 3 clearly shows that local demand varies between 6 and 14% of the total sales and, during the past years, local sales have decreased slightly. Compared with the local sales, the mill gained operating profit from the local market yearly from 10 to 15% of total operating profits. During the past years, the share of the locally based profit has slightly increased, which shows that the mill has lost its competitive- ness in the export markets. Generally, the home market effect in the Finnish paper deliveries has always had a minor impact on the mill’s sales and economy. The Finnish mills must adapt to serve long-distance customers, mainly printing houses, in foreign markets as cost-efficiently as possible. If paper demand should decrease in some export market, new markets and buyers must be searched for continuously in other foreign locations. The Finnish and other European paper markets are ma- tured, and demand will not increase very much, if at all (Hetemäki & Hänninen 2009; RISI 2009). In May 2005, there were strikes and block-outs in Finland, which caused large and severe produc- tion disturbances in the paper industry. Paper manufacturing experienced a total shutdown of several weeks. The foreign sales and even deliver- ies were significantly lowered or even stopped, which essentially decreased profits. The paper mills were only able to deliver normally to local customers. During these strikes, total profits from foreign sales dropped dramatically, and the Finn- ish mills may have lost customers to foreign mills during the incidence, even permanently. The peripheral location of the case mill can be easily illustrated by utilizing the transportation costs as presented in figure 4. The Finnish mills lack the clear and measurable cost advantage in transportation costs and delivery times of their lo- cal european competitors, which the Finnish mills must adapt to in some way, mainly with scale in production. Figure 4 implicitly supports the earlier studies (e.g. Hameri & Lehtonen 2001; Koskinen 2009); when a Finnish mill exports a large part of its sales, the paper products need local intermedi- ate storage, which produces high logistics costs. The export transportation costs (€/t) are averagely ten times higher than those for deliveries in the Fig. 3. Share of the home market in percent of total and the home market’s monthly share of total profit during 2001−2008. FENNIA 189: 2 (2011) 69Economic geographic characteristics in the Finnish paper … home market. This is one of the key challenges to be solved in the future. When examining how the mill’s economic per- formance has developed during the past years, three important values have been selected – vari- able costs (materials and energy), net profit and machine hours – to highlight how the economic efficiency has evolved (Fig. 5). This topic has been neglected in geographical studies earlier. Figure 5 clearly reveals that the mill’s economic perform- ance has decreased, costs per machine hour have increased and the mill has made less profit per ma- chine hour. It is highly alarming when economic productivity decreases. This development can also be presented from another perspective, when we look at how the main economic variables have de- veloped during the research period (Fig. 6). Aver- age paper prices have measurably lowered, while manufacturing and transportation costs have in- creased, lowering incomes. The case mill has not succeeded in compensating the decreasing prices by lowering manufacturing costs. The Finnish mills have obviously lost some part of their com- petitive advantage in paper markets. This econom- ic development (price–cost relation) has naturally been one of the main reasons for single machine lines, and even for large paper mills, having been closed in Finland and all over Europe. As many regional and economic geographical scientists from Weber (1909/1929) to Krugman (1995) have discovered earlier, our empirical data shows that at present, transportation and manufacturing costs have a remarkable influence on total costs and, later, inevitably on location in the paper industry. From the ‘Weberian’ perspective, transportation costs combined with manufacturing costs in- crease, and lowering paper prices together with decreasing economic performance, presented in figures 5 and 6, presumably will steer the mill lo- cation decisions made by paper companies in the future. Geographical distance affects economic varia- bles differently, as it is highlighted in figure 7. The figure presents the average variable and transpor- tation costs and sales prices to 73 export countries monthly during 2008. The data has been sorted by the estimated transportation distance (km) from Finland to the export countries. The figure shows that geographical distance makes the transporta- tion costs fluctuate heavily. Long sea routes do not seem to increase transportation costs as much as could be expected; in very long-distance deliver- ies, the average transportation unit costs are even slightly lowering. Figure 7 shows an interesting finding; a Finnish paper mill can partly compensate its disadvanta- Fig. 4. Average transporta- tion costs in €/t to export and home customers monthly during 2001–2008. 70 FENNIA 189: 2 (2011)Esa Hämäläinen Fig. 5. Development of the economic efficiency calcu- lated in euro per machine hour during 2001–2008. Fig. 6. Development of the average sales prices, variable and fixed costs and transport costs during 2001–2008 monthly. geous location from some of the distant markets with slightly higher sales prices. These empirical results partly support Krugman’s (1991) and Kil- kenny and Thisse’s (1999) ideas that companies should try to compensate their disadvantageous location with higher sales prices. The results show that the price differences between different coun- tries are large, up to over 30%. Figure 7 highlights that the local competition in foreign paper markets is country-specific with highly varying sales pric- es. In every market, there is a unique pricing situa- tion depending on the supply–demand equilibri- um. The exported paper qualities are usually so- called improved qualities, which may have a higher demand in foreign markets due to lower competition. The customers’ geographical posi- FENNIA 189: 2 (2011) 71Economic geographic characteristics in the Finnish paper … tion does not significantly increase the average transportation costs to distant customers, because sea transportation freights, even for heavy bulky products, are relatively inexpensive. In 2001, the average oil price was about 20 USD per oil barrel and in 2008, at the highest lev- el, about 150 USD per barrel (in 2007 prices). The effect of oil prices on shipping can be directly translated into increased bunker costs, whereby fuel costs represent as much as 50–60% of the to- tal ship operating costs, depending on the type of ship and services (TSA 2008; WTRG 2008). Figure 8 shows the distribution of the exported total tons in relation to average transportation costs to 73 different markets. Half of the paper tons (vertical line) from the case mill are delivered Fig. 7. Prices and transporta- tion costs (2008, €/t) to ex- port countries from the mill (N=73). Data was sorted by the estimated distance (km) from Finland to export coun- tries and distance rises from the left to the right. Fig. 8. Distribution of the ex- port transportation costs in €/t to 73 countries and cu- mulated delivered paper tons during 2008. 72 FENNIA 189: 2 (2011)Esa Hämäläinen mainly to 30 export countries, where the average transportation costs rise to over 90 €/paper ton (horizontal line). Competing in these markets by lowering obligatory transportation costs is not an easy task, especially due to oil prices, which di- rectly affect freight expenses. The lowering paper demand in Europe is a real challenge for the pe- ripherally located mills in the future due to trans- portation. Paper prices should increase, not lower, in the coming years. Figure 9 presents all the customer orders deliv- ered to three specific distant countries, namely USA, Australia and New Zealand (a total of 591 orders during 2001–2008). The research mill transports only 10% of its production to distant overseas countries. The data is sorted by customer- based deliveries in tons, which increase from left to right. The empirical data shows that transport costs have large variation and can rise threefold with the same volume of tons, naturally depending on the number of the intermodal (truck–sea–truck) routes. There does not seem to be a relation be- tween larger volumes in tons and lower transpor- tation costs, which could be expected. The case mill sells about 80% of its production in the European market; therefore, all large and population-rich countries are important for the mill. Figure 10 presents the statistical dependency between transportation costs and gross margins in the Spanish market. The higher transportation costs (r2= 0.2460, p<0.05) seem to have some sta- tistical dependency with lower gross margins. The high volatility in gross margins probably depends on the paper qualities and sizes of deliveries in paper tons (see e.g. Hämäläinen & Tapaninen 2010) and on the very different locations of the end customers in relation to the distribution cent- ers. The paper market is, according to RISI (2009), under severe competition in Europe, and the print- ing paper market is hardly growing at all. There is a lot of capacity and oversupply in the European market, which is affecting paper pricing. This makes reliable economic planning and estima- tions very challenging for the mills. The empirical material shows that the paper in- dustry is worth exploring with geographical mate- rial, and comparing the different countries can re- veal interesting and varying results in the econom- ic variables. This study supports the basic ideas of the economic geographers that there must be em- pirical studies concerning spatial topics to also present the views of real industrial manufacturing. The results of this study can be crystallized in that the economic performance of the case mill has de- clined in many areas. Location issues are expected Fig. 9. Relation between transport costs (€/t) and sales tons during 2001–2008. Data is sorted by the sales tons and are based on the all delivered customer orders (N=591) to distant countries (USA, Australia, New Zea- land). FENNIA 189: 2 (2011) 73Economic geographic characteristics in the Finnish paper … to become more central in the coming years due to demand and cost reasons. Discussion and conclusions In this paper, the paper industry is examined in a spatial context with empirical economic mill data. The topic of this article was to explore and describe how economic factors – like prices, eco- nomic performance per machine hour, transporta- tion and other costs – have developed. The paper industry has been examined somewhat marginal- ly in Finland from the viewpoints of both econo- my and geography together. The data was ob- tained from the economic system of a large mill, and the longitudinal data covers the years 2001– 2008. The original idea of this paper was particularly to adapt economic geography through empirical mill data, which seemed to give revealing results. Krugman (1995) argues that transport costs are a key topic in economic geography. Transportation costs, like other costs and paper prices, should be set by the market in the studies, as this examina- tion highlighted. This case study supports what e.g. Weber (1909/1929) and later Krugman (1995) have argued in their theoretical economics geo- graphical discussions: transportation costs affect economic results. The results show that location and economic geographical characteristics are significant for the paper mills’ results, like Arbia (2001) notes. The falling paper prices and rising costs were presented in figure 6, and in relation to this, the RISI (2009) has forecasted that this trend in the Eu- ropean paper demand and prices will continue. With the present oversupply situation (see e.g. RISI 2009), this may lead to paper companies closing their peripheral mills, simply because paper prices do not cover the high logistics and manufacturing costs. The situation is now more current than in the 1990s, when oil was inexpensive and paper demand and prices were higher. This study brings out interesting findings, which have distinct effects on the existence of the Finnish paper mills. These findings are prioritized below as follows: 1. Printing paper demand has lowered in the Euro- pean markets, and this increases oversupply and competition and presses down sales prices. 2. Manufacturing costs have increased during past years in the Finnish paper mills. 3. In exports, transportation costs are averagely ten times higher than in local deliveries, and these very high export costs give clear advantage to the local European competitors. Fig. 10. Correlation between transport costs and gross margins based on the month- ly data from 2001–2008. Data is sorted by the trans- portation costs euro/paper ton based on the paper deliv- eries to the Spanish market. 74 FENNIA 189: 2 (2011)Esa Hämäläinen 4. Economic performance in relation to machine hours in the older paper machines has lowered significantly during past years, which is an alarming signal. 5. Transportation costs have statistical dependency with gross margins by lowering them. 6. To distant overseas countries, transportation costs do not significantly rise due to relatively inexpensive sea transportation. The Finnish mills have obviously lost a lot of their ability to compete in the difficult European export markets and, as managerial implications, it can be considered that: • Paper companies should eliminate oversupply from the market to get paper prices up in order to cover the soaring material and transportation costs. • Methods to make logistics more cost-efficient should be examined carefully. These ongoing and quite rapid changes in the paper market can be forecasted to bring out severe challenges for the Finnish paper producers in the coming years. Further paper mill closures can be expected to balance the supply–demand equilib- rium. The large global paper companies are obvi- ously already considering relocating bulky and low-profit paper production from the periphery to more affordable places closer to the market in some scale. This study showed that the export-in- tensive paper industry is worth studying with em- pirical and geographical mill data to reveal the differences of markets. Further research The economic topics on country, customer and grade level are interesting subjects for future ex- aminations. An interesting but challenging subject would be to reveal basic issues affecting delivery and operating profit variations. Additionally, the sales and price estimations and actual deliveries and prices should be examined together in more detail in future studies to reveal the accuracy level of the forecasts. This could help in finding out how the Finnish mills have succeeded in estimating the demand and prices in different main markets. ACKNOWLEDGEMENTS This research was funded by Metsäteho Oy. REFERENCES Aiura H & Sato Y 2009. A model of urban demogra- phy. Discussion Papers in Economics and Busi- ness, 09–18. Osaka University, Graduate School of Economics and Osaka School of International Public Policy (OSIPP). Arbia G 2001. Modelling the geography of economic activities on a continuous space. Papers of Re- gional Science 80, 411–424. Arlbjørn JS, Jonsson P & Johansen J 2008. Nordic re- search in logistics and supply chain management: an empirical analysis. International Journal of Physical Distribution & Logistics Management 6, 452–474. Ballou R 2001. Unresolved issues in supply chain de- sign. Information Systems Frontiers 3/4, 417−426. Baxter JA & Chua WF 1998. Doing field research: practice and meta-theory in Counterpoint. Journal of Management Accounting Research 10, 69−83. Behrens K, Gaigné C & Thisse J-F 2009. Industry loca-Industry loca- tion and welfare when transport costs are endog- enous. Journal of Urban Economics 65, 195–208. Bullock K & Cliff D 2004. Complexity and emergent behaviour in ICT. Report commissioned by Fore- sight programme, OST, Hewlett Packard. Diesen M 1998. Economics of the pulp and paper industry. Gummerus, Jyväskylä. Dieter M & Englert H 2007. Competitiveness in the global forest industry sector: an empirical study with special emphasis on Germany. European Journal of Forest Research 126, 401−412. Eloranta E, Ranta J, Ollus M & Suvanto P 1994. Uusi teollinen Suomi. WSOY, Juva. Essletzbichler J & Rigby DL 2007. Exploring evolu- tionary economic geographies. Journal of Eco- nomic Geography 7, 549−571. Flyvbjerg B 2006. Five misunderstandings about case-study research. Qualitative Inquiry 12, 219−245. Forestindustries 2009. 8.11.2011. Forslid R, Haaland J, Midelfart I & Knarvik KH 2002. A U-shaped Europe? A simulation study of indus- trial location. Journal of International Economics 57, 273−297. Gallis C 1997. Stochastic computer simulation of for- est biomass logistics in Greece. Ph.D. disserta- tion, Department of Forest Resource Manage- ment, University of Helsinki. Hameri AP & Lehtonen J-M 2001. Production and supply management strategies in Nordic paper mills. Scandinavian Journal of Management 17, 379−396. Hetemäki L & Nilsson S 2005. Conclusions and Im- plications. In Hetemäki L & Nilsson S (eds). Infor- mation Technology and the Forest Sector 18, 221−233. IUFRO World Series. Hetemäki L & Hänninen R 2009. Arvio Suomen puun jalostuksen tuotannosta ja puunkäytöstä FENNIA 189: 2 (2011) 75Economic geographic characteristics in the Finnish paper … vuosina 2015 ja 2020. 10.5.2009. Hämäläinen E & Tapaninen U 2008. Spatial charac- teristics of the transports in the paper mill’s supply chain. Fennia 186, 83–93. Hämäläinen E & Tapaninen U 2010. Economics of a Nordic paper mill – case study. Industrial man- agement & Data Systems 110, 5−23. Hölsä J 2005. Metsäteollisuuden kuljetusratkaisut. Kymenlaakson Kauppakamari, Logistiikkapäivät, 30.5.2005. Isard W 1956. Location and Space-Economy. The M.I.T. Press, Cambridge, MA. Kilkenny M & Thisse J-F 1999. Economics of loca- tion: A selective survey. Computers & Operations Research 26, 1369−1394. Koskinen P 2009. Supply chain strategy in a global paper manufacturing company: a case study. In- dustrial Management & Data Systems 109, 34−52. Koskinen P & Hilmola O-P 2008. Supply chain chal- lenges of North-European paper industry. Indus- trial Management & Data Systems 108, 208−227. Krugman P 1991. Increasing returns and economic geography. Journal of Political Economy 99, 483– 499. Krugman P 1995. Development, Geography, and Economic Theory returns and economic geogra- phy. The M.I.T. Press, Cambridge, MA. Lamberg J-A & Ojala J 2006. Evolution of competitive strategies in global forestry industries: Introduc- tion. In Lamberg J-A, Näsi J, Ojala J & Sajasalo P (eds). The evolution of competitive strategies in global forestry industries: comparative perspec- tives, 1–29. Springer, Dordrecht. Lehtonen J-M 1999. Supply chain development in process industry. Acta Polytechnica Scandinavi- ca. Industrial management and Business Adminis- tration, Series No. 4, Espoo. Lewis MW 1998. Iterative triangulation: a theory development process using existing case studies. Journal of Operations Management 16, 455– 469. Lähtinen K 2007. Linking resource-based view with business economics of woodworking industry: earlier findings and future insights. Silva Fennica 41, 149−165. Martin R 1999. The new ‘geographical turn’ in eco- nomics: Some critical reflections. Cambridge Journal of Economics 23, 65–91. McCutcheon DM & Meredith JR 1993. Conducting case study research in operations management. Journal of Operations Management 11, 239–256. Oinonen H 2008. Challenging prospects for the Finn- ish paper industry. Paper and Timber 90, 14−18. O’Sullivan D, Manson S, Messina JP & Crawford T 2006. Space, place, and complexity science. En- vironmental and Planning A 38, 611−617. Ottaviano G 2003. Regional policy in the global economy: Insights from New Economic Geogra- phy. Regional Studies 37, 665−673. RISI 2009. Paper Cuts – Pricing it can only go one way, and that is down. 12.5.2009. Scott AJ 2004. A perspective of economic geography. Journal of Economic Geography 4, 479–499. Scott AJ 2006. Geography and Economy. Oxford University Press, Oxford. Tappi 2008. UPM to shutter Kajaani paper mill, Tervasaari pulp mill in Finland by year’s end. 10.5.2009. TSA 2008. 8.11.2011. Weber A 1909/1929. Über den Standort der Industr- ien. Tübingen. English translation by Freidrich CJ 1929. Alfred Weber’s Theory of Location of Indus- tries. University of Chicago Press, Chicago. WTRG 2008. Oil price index. 10.5.2009. Yin RK 2003. Case study research, design and meth- ods. 3rd ed. Sage, London.