Resources Policy 46 (2015) 127–138
Contents lists available at ScienceDirect
Resources Policy
journal homepage: www.elsevier.com/locate/resourpol
The employment and population impacts of the boom and bust of
Talvivaara mine in the context of severe environmental accidents – A
CGE evaluation
Hannu Törmä a,n, Susanna Kujala a, Jouko Kinnunen b
a
b
University of Helsinki, Ruralia Institute, Kampusranta 9 C, 60320 Seinäjoki, Finland
Statistics and Research Åland, Ålandsvägen 26, Åland, Finland
art ic l e i nf o
a b s t r a c t
Article history:
Received 24 February 2015
Received in revised form
9 September 2015
Accepted 9 September 2015
There had been a mining boom in Finland before the current recession. The most ambitious investment
was the Talvivaara nickel and zinc mine in Kainuu. The operation phase began in 2008, and for three
years the mine produced nickel and zinc according to expectations. Then everything changed: two accidents occurred in 2012, which had severe environmental consequences. There was a failed attempt at
corporate restructuring. The production company of Talvivaara is now in bankruptcy, and the national
government is financing the mine. Our aim is to present an evaluation of the impact these events had on
the employment and population of Kainuu region. Our results for the period 2009–2014 indicate that the
Talvivaara mine still had a positive cumulative effect on the employment of Kainuu, in spite of the environmental accidents. The results for the period 2015–2022 suggest that the full implementation of the
rejected corporate restructuring plan would have been a tolerable solution for the employment and
population of Kainuu region. Considering the uncertain future of the mine, we suggest follow up studies.
& 2015 Elsevier Ltd. All rights reserved.
JEL classification:
C68
J11
R11
Keywords:
Economic impacts
Mining
CGE modeling
Evaluation
1. Introduction
There had been a mining boom in Finland prior to the current
recession. The prevailing expectations are that a growing mining
sector could compensate for some losses in economic growth and
employment, whereas the traditional export industries metal,
pulp, paper and wood industries have suffered during the recession. The most sparsely populated regions in East and North Finland, which are the target regions of most new mining projects,
might yet gain some advantage from mining.
The most ambitious investment was the Talvivaara nickel and
zinc mine in Kainuu. The operational phase of the mine began in
2008 and for three years the mine produced nickel and zinc in line
with expectations. Then everything changed, two accidents occurred in March and November 2012, which led to severe environmental consequences. This lead to increasing environmental
protection and rehabilitation costs, losses in production, financial
problems, and subsequently the mine applied for corporate restructuring. The restructuring plan required the large downscaling
n
Corresponding author.
E-mail addresses: hannu.torma@helsinki.fi (H. Törmä),
susanna.kujala@helsinki.fi (S. Kujala), [email protected] (J. Kinnunen).
http://dx.doi.org/10.1016/j.resourpol.2015.09.005
0301-4207/& 2015 Elsevier Ltd. All rights reserved.
of the company's huge debt. The creditors, however, did not accept
the plan, and the production company in Sotkamo Kainuu went
into bankruptcy. The government of Finland is currently financing
the mine, trying to rectify the remaining environmental threats,
and also find a new owner for the mine.
The purpose of our study is to evaluate the employment and
population effects of the Talvivaara mine for the Kainuu region
over two time periods. The first period covers the years 2009–2014
for which all economic data are available. The second period is
projected for the years 2015–2022. This scenario is based on the
full implementation of the plan of corporate restructuring. Our
evaluation tool is the dynamic general equilibrium model, CGE
RegFinDyn of the Finnish economy. The elements of the various
scenarios include the actual or assumed production level, mining
and environmental investments, and the debt burden.
First, we present figures that describe the mineral resources of
Finland and the mining industry. We also describe the economy of
the Kainuu region. Subsequently, we outline some methods that
can be used to evaluate such mining projects. The Talvivaara
mining project has been detailed, and an account is given of the
environmental accidents that occurred at Talvivaara. The plan for
the corporate restructuring is also recounted. Then the RegFinDyn
model, the assumptions for the scenarios and the employment and
population effects for the Kainuu region are presented. The
128
H. Törmä et al. / Resources Policy 46 (2015) 127–138
conclusions, lessons learned and policy implications are presented
at the end of the text.
2. The Finnish mining industry
Finland is rich in minerals (Figs. 1 and 2). There are 13 metal ore
mines that are currently operating or are under construction or are
in the planning phase. Kittilä, Laiva and Pampalo gold mines and
Talvivaara nickel and zinc mine are already operating. Kevitsa
nickel and copper mine began production in 2013.
Apart from its diverse mineral potential, Finland provides a
good operating environment for exploration and mining activity.
Fig. 1. Industrial mineral and rock mines and quarries of Finland.
Source: Geological Survey of Finland (2009).
Recent international estimations have even suggested that Finland
is one of the most favored countries for targeting mining operations at (Geological Survey of Finland, 2010). Finland has just
amended its mining legislation. The new Mining Law (621/2011)
not only creates a favorable environment to securing the preconditions for mining and ore prospecting, it also covers environmental issues including: citizens' fundamental rights, landowners' rights and municipalities' opportunities to influence decision-making.
When metal prices were low, Finnish firms sold the mining
rights to foreign firms in most cases. The mining boom in Finland
preceded the economic downturn. Exploration began to grow,
especially after world market prices for metals increased.
H. Törmä et al. / Resources Policy 46 (2015) 127–138
129
Fig. 2. Metal concentration mills, and producers of steel and base metals in Finland.
Source: Geological survey of Finland (2009).
Table 1
Finnish metal mining industry.
Comparison/year
2007
2008
2009
2010
2011
2012
2013
Number of enterprises, and (locations)
Production of metal concentrates, million tons
Personnel
Turnover, € million
Exports, € million
20 (25)
1.2
737
179.7
21.9
17 (26)
1.3
1058
141.7
11.8
18 (27)
0.7
1071
158.0
3.1
22 (30)
1.4
1246
379.4
31.9
21 (28)
1.7
1669
627.8
57.2
21 (29)
1.7
1949
700.2
88.6
26 (27)
2.4
1993
685.1
126.5
Sources: Kokko (2014), Database Industry Online (2013).
130
H. Törmä et al. / Resources Policy 46 (2015) 127–138
Kainuu due to the Talvivaara mine. Mining employed 185 persons
in the Kainuu region in 2005, and that total rose to 705 in 2012.
However, the numbers of employed persons in other primary
production and manufacturing sectors in Kainuu fell between the
2005 and 2011 period. Similar changes also happened in the whole
Finland, but the proportion of employed persons in the manufacturing sectors in Kainuu decreased a lot more than for the
whole of Finland. In contrast, the proportion of the work force
employed in the service sectors increased in Kainuu during the
same period. This follows the same trend as for the whole of
Finland.
A CGE model is an equilibrium model, therefore there are
usually both winners and losers at the sectoral level using such a
model. The possibility of a crowding out effect, from mining is
built into the model as an intrinsic constituent. The simulation
results for the period 2009–2012 show how the mine had affected
the employment of other primary and industrial sectors. There is
one primary production sector, forestry that lost employment to
mining and other sectors serving the mining sector in the 2009–
2012 period. The losses in employment in the industrial sectors
were in the following: the processing of foodstuffs, the manufacturing of textiles, the pulp and paper industry, and in the
manufacturing of electric equipment and vehicles. The employment losses range from 0.1% to 1.9% per year, and when summed
over a four-year period ranged from 0.3% to 7.6%. Our interpretation is that the Talvivaara mine had a crowding out effect, but this
effect was modest in size and happened in only a few primary
production and industrial sectors.
The total number of employed persons in Kainuu increased
between 2005 and 2012. The unemployment rate in Kainuu fell
significantly between 2005 and 2011. In 2011 the unemployment
rate at 8.3% was exactly half of the rate that was in 2005. The
unemployment rate in 2012 went up to 11.4% that is near the level
in 2008. This figure was clearly affected by the problems at the
Talvivaara mine, which began during the same year.
Talvivaara nickel and zinc mine was one of the earliest projects in
the boom, thus its investment phase started in 2007, and its production phase began in 2008. Some basic statistics from the Finnish mining sector are presented in Table 1.
The number of enterprises has not increased by much, but
there has been some expansion in certain locations. The same is
true for the production of base metal concentrates for further
processing. The numbers of personnel of these mining sites have
increased substantially, which indicates that there are several
mining investments that are in the construction phase. The turnover has also increased greatly because some new mines had already begun to operate. This expansion in mining has fostered
exports.
3. The Kainuu region
The region of Kainuu is in eastern Finland. The Talvivaara nickel
and zinc mine lies in Sotkamo, a municipality that is located in the
southern part of Kainuu region. The size of Kainuu's population is
currently 81,000. Kainuu is a region whose population has been
declining for decades. The population has been decreasing by at
least 500 persons per year, since 2005. Kainuu's GDP increased
steadily from 2005 to 2008. Then the GDP of Kainuu region decreased by almost €300 million between years 2008 and 2009. The
economic depression in Finland, which is still continuing began
during the last quarter of 2008. Finland's national GDP decreased
by 8.5% during 2009. Kainuu was among the worst affected Finnish
regions. The industry sector suffered the most, its turnover decreased by over €300 million between the years 2008 and 2009
(Database industry online, 2013). As a consequence the regional
GDP of Kainuu decreased by 12.0% during 2009. The depression is
the cause behind the decrease in Kainuu’s economic growth. The
recovery of regional GDP during 2010–2012 was due to the mining
investments in Talvivaara and their multiplier effects for instance
in industry and services sectors. The regional GDP of Kainuu returned to about the same level as in 2008 (pre-downturn) during
the 2010–2012 period.
Mining investments have increased notably from 2006 to 2008
by which time they had reached a peak of over €400 million as
shown in Table 2. Investments in other sectors have been steadier,
although service sector investments have slightly increased between 2005 and 2012. The total investments in Kainuu increased
from €407 million in 2005 to €589 million in 2012 mainly because
of the mining investments in Talvivaara.
Mining has become an increasingly important employer in
4. Mining as an evaluation topic
There is a multitude of quantitative evaluation studies about
infrastructure investments in large scale activities such as in
mining. It is impossible to cover all of the literature, but we will
mention some recent research and focus on the evaluation
methods used. Our references describe the uses of input–output
(IO), social accounting matrix (SAM) multiplier and non-multiplier
CGE models. It is widely accepted that the evaluation results of the
Table 2
Kainuu region's population, GDP, employment and investments between 2005 and 2012.
Variable/year
2005
2006
2007
2008
2009
2010
2011
2012
Population
Change in population from previous year
GDP, € million
Investments, € million
Mining
Other primary production
Manufacturing
Services
Total
Employment, numbers of employed
Mining
Other primary production
Manufacturing
Services
Total
Unemployment rate, % of labor force
85,634
635
1786
84,827
808
1929
84,065
762
2005
83,470
595
2117
82,897
573
1864
82,354
544
2044
81,686
668
2175
80,992
694
2182
12
48
56
291
407
6
49
78
276
409
151
50
53
346
600
403
56
88
320
866
171
59
72
286
588
150
49
43
285
528
284
50
63
327
724
136
46
48
360
589
185
3722
6529
23,712
34,149
16.6
218
3712
6800
23,927
34,657
17.1
292
3802
6838
24,349
35,281
15.7
434
3682
7032
25,010
36,158
11.2
563
3092
5800
25,069
34,523
9.3
544
3151
5616
25,769
35,079
9.0
647
3077
5561
25,798
35,083
8.3
705
3150
5534
25,634
35,023
11.4
Source: Statistics Finland (2014).
H. Törmä et al. / Resources Policy 46 (2015) 127–138
IO models represent the upper limits, thus they tend to overestimate the effects, of an investment. The IMPLAN1 model (see
www.implan.com) combines the approaches of the first two
methods. The database in the CGE approach typically include values obtained from the IO tables, which is a good way of describing
the money flows between the firms, consumers, regional and national governments. Ejdemo and Söderholm (2011) conducted a
survey of selected mining studies. Rolfe et al. (2011) provide a
good introduction to the the IO method.
Ejdemo and Söderholm (2011) used a regional IO model, rAps,
to research large scale iron ore mine project in Pajala, Sweden.
According to their results, a typical mining employment multiplier
lies in the 2–2.5 range. Consequently, one working place at the
mine would be expected to create 1–1.5 additional jobs elsewhere
in the economy. Weisskoff (2010) studied the Kings Road limestone mine project in Levy County USA by using the IMPLAN
model. The results indicate that the employment effect would be
even larger in the hunting and fishing sector than in the mining
sector, even if the sectors were to have the same amount of economic activity. Rolfe et al. (2011) studied the Codrilla coal mine
project in Queensland Australia using an IO model and tables
generated by the GRIT method/model. According to those authors,
the positive economic impacts of the project on the regional
economy need to be accompanied by a focus on managing potential housing and labor force impacts. An investigation conducted by Skurla et al. (2009) used the IMPLAN model to study
ferrous and non-ferrous mining in Northeast Minnesota USA and
determined an employment multiplier in the range of 2.5–2.8.
Peach and Popp (2008) also used the IMPLAN model to research
new uranium mining and milling operations worth of $1.1–2.5
billion, in New Mexico. According to their study, the base case
scenario gives a total employment effect of 12,586 jobs, 6921 of
which was the direct effect of mining.
CGE models are well-suited for the analyses of changes in
economic conditions that have wide-spread economic effects. CGE
models efficiently serve as evaluation tools in the analysis of
mining investment cases. The economic effects of mining are
wide-spread, so a partial picture of their effects is just not enough.
CGE models are based on established micro and macro-economic
theories and well-established methods of econometrics and applied mathematics. The models were developed to overcome
weaknesses of the older linear techniques. They take into account
the fact that the decision making rules of economic actors can be
or are non-linear. They can also account for resource restrictions,
such as a limited pool of labor and insufficient capital or land in a
region. The modeling of public sector incomes and expenditure are
included in these models. Reactions to changes in domestic and
foreign trade and investments are modeled. Prices are endogenous
and serve as mechanisms for adjustment towards a new equilibrium after a change in economic conditions. Nowadays, CGE is
the mainstream type of methodology that is used for the analysis
of economic effects that result from private and public investments or for evaluating the impact of policies. There is a survey of
several national and regional CGE models published by Törmä
(2008) and also by Rutherford and Törmä (2010).
Researchers at Centre for International Economics (2010) examined the Sepon copper and gold mine of central Laos, by using
an economy wide CGE ORANI-G model. Their results suggested
that since the start of production, 2003/2005 the mine had contributed up to 5.6% of the entire GDP of Laos. In addition to this
result, indirect effects amounting to 2.6%, were also generated
which took the total effect to 8.2%. Ye (2008) used a dynamic CGE
MMRF-Green model to study the likely impacts of the commodity
1
IMPLAN comes from words: Impact Analysis for Planning.
131
boom in the iron ore sector on the Western Australian economy.
According to his results, the economy was expected to benefit
from the expansion in iron ore exports and from associated investment in terms of rising consumption and employment, although at the industry level there will be winners and losers. The
results of the study by Wittwer and Horridge (2010) suggest that
the terms-of-trade boom in Australia brought with it accelerated
investment in the mining industry. This boost led to huge jumps in
house prices in some of the mining regions. The four year construction phase was found to increase the housing stock price by
as much as 30%. Wittwer and Horridge also studied regional impacts of rising demands for raw materials in China and India, plus
the effects of the recurring droughts in south-eastern Australia by
using the CGE TERM model.
Castillo and Yu (2014) used a CGE model to study impacts of
liberalizing the Philippine mining industry. According to their results, the Mining Act of 1995, and especially the part that liberalized investments, has negative welfare implications on households
and some sectors. The Mining Act of 1995 gave foreign investors
the right to have full ownership of a mining enterprise instead of
previous 40%. Bohlmann et al. (2014) measured the economy-wide
impacts of the labor strike in South Africa's platinum industry in
2014. The analysis was made using the dynamic CGE model of
South Africa. Those authors found that the real GDP growth decreased by at least 0.7% because of the strike. They also found out
that the higher nominal wages was not the most damaging outcome of the strike, because it seemed to affect the economy only
temporarily. The investors' possible reactions in the mining sector
towards South Africa could have longer lasting negative impacts
on the South African economy. Downes et al. (2014) have estimated the impact of mining boom on the Australian economy.
They used a large-scale structural macro econometric model AUSM, which is modern Keynes-Klein-style model, to which CGE
features have been added. Their results suggest that the mining
boom increased real per capita household disposable income by
13% by 2013.
5. Finnish mining investment studies
The University of Helsinki, Ruralia Institute conducted a total of
15 case studies of mining investments over the 2007–2014 period.
The reports are available at: http://www.helsinki.fi/ruralia/re
search/regfin.htm. Most of these studies were on individual
mining investments, and covered both the construction and operation phases of the mine. There are two larger studies for several
cases. The study by Törmä and Reini (2009) presented a comprehensive analysis for nine new mining investments, which were
either in the planning or construction phase. Laukkonen and
Törmä (2014) conducted an update study for the same mines and
one additional new mining investment. Six of these 10 mines are
now operating, and four are in the planning phase. The Talvivaara
zinc and nickel mine is represented in both research studies.
The evaluation results of Laukkonen and Törmä (2014) indicate
that the mining sector in Finland has been efficient in the creation
of additional employment especially in the East and North of
Finland. According to their calculations, the six mines already
operating have created annually a mean of 1949 person working
years over the period 2009–2013. It was estimated that the corresponding figure could be 1862 person working years for the
future period 2014–2020. The planned four new mines would
contribute a mean of 2815 person working years annually over the
first 10 years.
132
H. Törmä et al. / Resources Policy 46 (2015) 127–138
6. The Talvivaara nickel and zinc mine2
Talvivaara Mining Company described itself as an internationally significant base metals producer that has a primary
focus on nickel and zinc production. Talvivaara's main asset was
the Talvivaara nickel and zinc mine, which is located in Sotkamo,
Kainuu Finland. The Talvivaara polymetallic ore comprise one of
the largest known nickel sulfide resources in Europe. Talvivaara
has an estimated 1121 million tons of ore as indicated by measured
and indicated categories, which is a quantity that is sufficient to
support the anticipated production for several decades. Production
at the mine started in October 2008 with the precipitation of the
first metal sulfides. The planned annual nickel output was 30,000–
50,000 tons and as by-products the mine would also produce
annually approximately 90,000 tons of zinc.
Talvivaara mine supplied metal intermediaries to companies
with metal refining operations and entered into a 10-year agreement with Norilsk Nickel Harjavalta Oy Finland for the entire
output of the mine's nickel and cobalt production at market prices.
In January 2010, Talvivaara entered into a long-term agreement
with Nyrstar NV Belgium for the latter to receive the concentrate
streaming of Talvivaara zinc output. Specifically, Talvivaara would
deliver all of its zinc in concentrate production to Nyrstar until
1,250,000 tons has been delivered.
Talvivaara uses the bioheapleaching process to extract the
metals from ore. The group first demonstrated the viability of this
novel technology by conducting large on-site pilot trials using the
Talvivaara ore as the stock material. Since July 2008, bioheapleaching has been used on production scale at the mine. The
mining company stated that leaching process has been shown to
be exothermic and therefore suitable for the sub-arctic climatic
conditions of Eastern Finland. They also said that the Talvivaara
ore body is well-suited for open pit mining due to a thin overburden, favorable resource geometry and a low waste to ore ratio.
Further, the ore is relatively low grade, but well-suited to bioleaching due to its high sulfide content. Currently, there are many
doubts as to whether the bioheapleaching technology is manageable in cold and rainy weather conditions3. The Talvivaara mine
has only been able to produce annually a maximum of 16,087 tons
of nickel, whereas 30,000–50,000 tons was the set target.
6.1. The environmental accidents4
In March 2012, a pipe fracture occurred during primary heap
reclaiming, which caused a discharge of approximately 10 cubic
meters of process solution onto ground that had not been covered
with a ground protection film. The solution accumulated in a
frozen ditch situated at the edge of the heap area. Another leak
occurred in the primary leaching solution channel. The process
solution was able to flow via a ditch into the after-treatment basin.
The effects of both leakages were reflected in exceptionally high
metal concentrations in Lake Kivijärvi in the spring, after which
the metal concentrations recovered to normal levels. There was a
third incident in March 2012 whereby approximately 280 cubic
meters of process solution flowed into the area between two
2
This section is based on the information from the www pages of the Talvivaara mining company (see www.talvivaara.com), and from the annual reports of
the company.
3
The authors are not aware of any scientific study on the limitations and inefficiencies of the bioheapleaching technology: especially under colder wetter
weather conditions and therefore such mal-functioning has not been scientifically
proven. What doubts that have been expressed about the efficiency of bioheapleaching in such conditions have only been articulated as the opinions of several
special interest groups and experts as editorials in leading newspapers and media.
4
Due to the specific terminology of mining, this section is a modification from
the information presented in the annual report 2012 of the Talvivaara mine.
solution basins onto ground that was not covered by the ground
protection film. The solution accumulated in a recess in the frozen
ground. In these cases most of the process solution was successfully removed.
In November 2012, there was a severe leakage in the gypsum
pond, which resulted in a suspension of the production of the
metals recovery plant. The staff concentrated their efforts on locating and blocking the leakage and on attempting to prevent its
adverse environmental impacts. As a result of this leakage, approximately one million cubic meters of acidified water containing
metals and their sulfates flowed into the after-treatment basins
that also serve as security catchment basins for the mine area.
During the initial stage of the accident, approximately 20,000
cubic meters of effluent water from the pond flowed into the
water discharge routes. During the accident, around 216,000 cubic
meters of acid and metal-containing waters had to be run through
the water discharge routes.
After these environmental accidents, the leakage from the
gypsum pond was located and stopped. The nearest water bodies
were neutralized. An overland flow area between the Kortelampi
pond and Lake Ylä-Lumijärvi was constructed. The objective was
to reduce the amount of metals and anions that were contaminating the water bodies. These actions demanded immediate
environmental protection and rehabilitation investments and slashed the production of metals and consequently the mine's turnover. The accidents led to many complaints from local inhabitants
and much distrust directed at the Talvivaara mine, and towards
mining in general. Currently four top managers, including the CEO,
of the Talvivaara mining company are facing court proceedings to
answer accusations of criminal neglect and causing damage to the
environment.
6.2. The corporate restructuring bid failed
Talvivaara mine applied for corporate restructuring towards the
end of 2013 due to the continued deterioration in its financial
position. As a result of the financial consultations, Talvivaara implemented the gradual lay-off of about half of its employees. A
corporate restructuring program was published in November 2014.
The starting point of the restructuring program was the sale of
all of Talvivaara Sotkamo's business operations through a realization restructuring process to a new company that was to be established by Talvivaara. The existing product streaming and product sale and purchase agreements that were concurrently held in
Talvivaara Sotkamo's name were to be transferred to the new firm.
The plan was that a one-off payment would be made to the
creditors with the possibility of making supplementary payments,
under an eight-year restructuring program provided that financing
for the duration of the implementation of the restructuring program could be secured. No payments would be made to creditors
of Talvivaara during the first two years of implementing the accepted restructuring plan. Thus, the payments to creditors were to
have taken place during the 2017–2022 period and that the
creditors were to be paid 10% of the capital cut in accordance with
the plan annually for the first two years (2017–2018) and 20%
thereafter (2019–2022). The secured debts of Talvivaara were to be
settled according to the same schedule as for the unsecured debts.
It was estimated in the plan that the sum of all the secured
restructuring debts of Talvivaara Sotkamo (in aggregate €130
million) that constitutes a financing debt of €53 million after the
deduction of liquidation costs. The plan proposed that €21.9 million of these secured debts was to be payable upon the execution
of the realization restructuring process. The secured creditors and
the parties to the sale and purchase of Talvivaara Sotkamo's
business were to agree separately on how the remaining balance
of the secured financing debt (€31.1 million) was to be paid. This
H. Törmä et al. / Resources Policy 46 (2015) 127–138
arrangement required the consent of all of the secured creditors. It
was estimated that the debt secured by a floating charge of Talvivaara was €7.5 million and the debt secured by other securities
was €3 million after the deduction of liquidation costs.
The plan proposed that the capital of unsecured debts of the
Talvivaara Sotkamo production company (in aggregate not less
than €956 million) and the Talvivaara mother company (in aggregate €478 million) was to be cut by 97–99%. No payments were
to be made on debts with the lowest priority of either of the
companies. The convertible bonds and bonds issued by the company were to be treated as an unsecured debt. The new company
established by Talvivaara was to have only €65.7 million of debt,
which was to be paid to creditors during the 2017–2022 period.
The creditors, however, did not accept the plan, and the production company in Sotkamo Kainuu went into bankruptcy in
November 2014. The national government is currently financing
the mine, trying to tackle the remaining environmental threats,
and also find a new owner for the mine.
7. The RegFinDyn model
The Ruralia Institute developed national and regional multisector and interregional CGE models over a 10 year period. The
national model is called GemFin (General equilibrium model for
Finland), for a description see Törmä and Lehtonen (2010). The
regional models are called RegFin5. The regional models are bottom-up models in which the regional effects of a change in economic conditions are solved for all regions simultaneously. National macro results for economic growth, employment, income,
consumption, tax revenues, public services and trade etc. are calculated alongside the regional results.
CGE RegFin models are available both as comparative-static and
also as dynamic versions. Descriptions of the comparative-static
model were presented by Törmä (2008), Rutherford and Törmä
(2010), Törmä et al. (2010), Törmä and Zawalinska (2010, 2011).
There is also a Pan-European model RegEU for the 27 European
Union member countries (LSE Enterprise, 2011). Since the comparative-static model is well described in the publications above,
we will focus on the dynamic model used here. There are two
additional modules for the capital and population dynamics.
7.1. Capital dynamics
The dynamic model version of CGE RegFin takes into account
the time dimension. Mining investments are typical examples of
projects that demand dynamic analysis. The reason is that in the
larger projects the investment costs materialize over several years.
The investment scenarios are calculated for a sequence of years.
The effects of a change in economic conditions can then be followed year-by-year or cumulatively. This kind of modeling represents recursive dynamics6 whereby the results for selected
variables for the current year are dependent on values of some
variables from the previous year. For example, an investment undertaken in year t is assumed to become operational at the start of
year t þ1. For the capital in sector i and region r, this is described in
133
the following accumulation relation.
Ki,r, t + 1 = ( 1–δ i,r ) x Ki,r, t + Yi,r, t ,
(1)
where
The left hand side of the equation define the quantity of capital
available at the start of the year tþ 1. Kt is the quantity of capital
available at the start of year t, δ is a fixed depreciation rate, and Yt
is the quantity of new capital created during year t. Given a
starting point value for capital in t ¼0, and with a mechanism for
explaining investment through time, this equation can be used to
trace out the time paths of sector capital stocks.
Investment in sector i in region r in year t is explained via the
following formula.
Ki,r, t + 1/ Ki,r, t – 1= Fi,r ( erori,r, t ),
(2)
where the left hand side defines the change of the capital stock,
eror is the expected rate of return on investment, and F is an increasing function of the eror term with a finite slope.
Investors are assumed to have static expectations, so they only
take account of current capital rents and asset prices when
forming current expectations about rates of return. The mining
investments will be made only if the rate of return to capital is
high enough to bring in sufficient amounts of income. The change
in economic conditions change the rate of return for capital invested. New investments are directed to those sectors and regions
that become more profitable than the others.
The percentage change in the expected equilibrium rate of return to capital, eerori,r is defined as follows.
eerori,r = ( 1/K slope, i, r ) x ⎡⎣ 1/( K gr,i,r – K grmin,i,r) + 1/( K grmax,i, r – K gr,i,r) ⎤⎦ x
(3)
dk gr,i,r,
where Kslope is the overall sensitivity of capital growth to
changes in the expected rate of return for capital, Kgr is the capital
growth rate between the start and end of year, min and max refer
to its lower and upper boundaries, and dkgr is the percentage
change in capital growth rate. The investment function is thus a
constrained logistic function. Setting the boundaries is a practical
way of securing reasonable scenario results for the investments.
The setting is presented in Fig. A1, where rorni is a coefficient
that represents a sector's historically normal ror, and trendKgri is a
coefficient, set to sector’s historical normal rate of capital growth.
For a sector to attract sufficient investment in year t for its
capital growth to exceed its long-term average, its eeror must be
greater than its normal level. Conversely, if the eeror on the sector’s capital falls below the normal level, then investors will restrict their supply of capital to the sector to a level below that
required to sustain capital growth at the rate of the trend level.
The capital depreciation rates represent the lower boundary
and a typical long term growth rate serves as an upper level. This
entails that the replacement investments are being done by assumption. If a change in economic conditions increases investments in net terms compared to the previous year in some sector,
the regional economy has more capital inputs in its disposal for
the next year.
7.2. Population dynamics
5
The model has been replicated earlier for Poland (Zawalinska, 2009). In the
EU funded FP7 CAPRI-RD project (see http://www.ilr1.uni-bonn.de/agpo/rsrch/ca
pri-rd/caprird.htm) a layer of single-country RegFin style models has been created
for all 27 European Union member countries by using NUTS 2 level data. The layer
of CGE models is linked to the partial equilibrium CAPRI model of agriculture (Britz,
2012).
6
Capital dynamics in RegFinDyn are basically the same as those used in the
Australian MMRF model (Monash Multi-Regional Forecasting Model, Adams et al.
2010).
The central equation, presented here in levels, for population
dynamics is as follows.
POPUa,s,r, t + 1 = POPUa + 1,s,r, t + BIRTHSs,r − DEATHSa,s,r
+ NINMIGa,s,r + WINMIGa,s,r − NOUTMIGa,s,r
− WOUTMIGa,s,r
(4)
134
H. Törmä et al. / Resources Policy 46 (2015) 127–138
where:
a ¼age 0, 1, 2 …, 95þ
s ¼gender
r ¼region
t¼ time, used where inevitable for clarity
POPUa þ 1,s,r,t
BIRTHSs,r
DEATHSa,s,r
NINMIGa,s,r
WINMIGa,s,r
NOUTMIGa,s,r
WOUTMIGa,s,r
Population in year t by age cohort, gender and
region made one year older,
0-year-olds by gender and region,
Number of deaths by age cohort, gender and
region,
Persons moving in from other Finnish regions
by age cohort and gender,
Persons moving in from abroad by age cohort
and gender,
Persons moving out to other Finnish regions
by age cohort and gender,
Persons migrating outside Finland by age cohort and gender.
Fertility levels by age of mother and by region are assumed to
follow the assumptions of Statistics Finland's latest population
forecast (2012). In other words, fertility rates are five-year cohortwise regional averages that are assumed to be constant over the
simulation period. Gender distribution of the newborn is assumed
to follow the national average for every region. The number of
newborn is thus dependent on the age structure of women in
fertile age. Mortality rates are also based on the Statistics Finland's
population forecast. Exogenous, individual trends are imposed to
each cohort, gender and region. The number of deaths is thus
dependent on the development of mortality rates and the age
structure of the population.
The treatment of migration makes the population outcome
endogenous, and different from Statistics Finland's forecast, if not
imposed to follow it under baseline. The main postulation is that
the increase of employment opportunities will also increase migration into the region r, if the difference between the regional and
the national unemployment rate becomes more favorable for the
region r. This regional unemployment differential compared to the
whole country will determine the development of in migration
flows. Another important assumption is that the age and gender
structure of in migrants is assumed to follow roughly the average
structure of in migrants recorded for the region in recent statistics.
The out migration flows are assumed to follow constant migration rates, estimated for each age cohort by gender for each
region. The out migration flows are then scaled with nation-wide
shifter variables that ensure that the sums of national in and out
migration flows are equal by age cohort and gender, as they must
be by definition. The age structure is another determinant of migration flows. Regions with few young persons do not thus have
extensive out migration, as persons around 20–30 years are the
most probable movers, so if there are so few of them the out
migration figures will be correspondingly low. The corresponding
unemployment rate vs. in migration elasticity value, which governs the sensitivity of migration flows to labor market situation, is
set to 0.05. This parameter was estimated from a panel of regional
migration and unemployment data of Statistics Finland. If the
unemployment differential changes in favor of the region r by one
percentage unit, then its in migration would increase by 0.05%.
Migration abroad is assumed to follow migration rates that
tally with the Statistics Finland's forecast. In migration also follows
the recorded five-year averages calculated by Statistics Finland's
forecasts. The national net migration level is set to follow the
assumption of Statistics Finland's forecast. However, in future applications, this assumption could be revisited and set to follow the
growth of the Finnish economy. However, the model does not
describe the economic situation in the rest of the world, thus even
such a nationally determined solution would be unsatisfactory.
Finnish history is ripe with examples how factors other than
economic factors have governed the international migration flows
to and from Finland. The allocation of international in migrants is
assumed to follow past year’s allocation trends, but it is also affected by the population growth rates of each region. Fast-growing
regions will get a larger share of newcomers.
7.3. Labor supply, wages and the unemployment rate
Labor supply is assumed to depend on the changes in the size
of age cohorts by gender. Thus in this version of the model, the
labor supply does not react to changes in economic incentives. The
real wages are growing around a national average growth rate. The
movement around the trend growth is affected by percentage
point changes in the unemployment rate of the region with an
elasticity of 0.5. Thus, the wage growth reacts to changes in
unemployment rate rather sluggishly. Therefore, regional differences in unemployment rates are quite persistent, as in the real
world. The unemployment rate then becomes just a comparison
between labor supply and demand, which are determined elsewhere in the model code.
7.4. The baseline of the Kainuu region
There are many organizations, research institutes, ministries,
banks, lobbyists, that publish forecasts for GDP, employment etc.
at the national level. Only recently, work to create national and
regional baselines began (Honkatukia and Ahokas, 2012). Consequently, the availability of regional economic forecasts is restricted
as yet. We decided to use a simple baseline because of these
limitations. The economic development of the Kainuu region was
tied to the following variables: national GDP, productivity of labor
and real wages. The realized values for these variables are available for the years 2009–2014.
GDP decreased a lot during the year 2009. The next two years
showed much better development. The depression continued
during the three next years, thus the net overall growth of GDP for
the three years was negative. We assumed that the depression will
gradually end over the years 2015–2017. Many Finnish economists
expect that the growth rate of labor productivity will fall from the
present level 2.0% to 1.5% in the near future. The growth of GDP
was assumed to be equal to the assumed labor productivity or 1.5%
for the years 2019–2022. We also assumed that the labor productivity abroad is equal to 1.5%. Finally, we assumed that real
wages will grow by 1.9% annually. This has been a fact in the past,
and is explained by the power of the labor unions. Real wages tend
to grow faster than labor productivity.
7.5. The closure
Under base scenario, the national GDP growth rate is determined exogenously, as previously explained. Economy-wide
productivity growth of labor is used as the endogenously adjusting
variable that fulfills the national GDP growth path. The regional
GDP figures are left to adjust endogenously to the national GDP
growth rate. Under the policy scenario7, GDP growth is
7
The policy scenario refer to alternative simulations, such as have been run in
the article for the years 2009–2014 (Section 8.1) and for the corporate restructuring, years 2015–2020 (Section 8.2).
H. Törmä et al. / Resources Policy 46 (2015) 127–138
endogenous and productivity growth exogenous. Apart from the
endogenous component of the technical progress, even an exogenous term is added to base scenario, and reflects an overall
trend in labor productivity. The same size of change is added to
import prices to reflect technical progress in the rest of the world.
Population growth rates by region are exogenously determined
until 2012, hence the regional population figures are endogenously determined by nativity, mortality and net migration.
The national population is then determined by the sum of regional
populations. However, the sum of international net migration each
year follows exogenously the Statistics Finland's forecast.
Land supply is fixed during the whole simulation period. Capital stocks follow the dynamics explained above. Wages growth is
assumed to follow an exogenous growth variable, 1.9% per annum
together with the assumed unemployment rate elasticity of wages
of 0.5. Labor supply growth is according to changes in the size of
age cohorts. The participation rate of each age cohort follows what
was recorded in the available statistics until 2010. Thereafter, the
participation rates for each cohort are fixed.
8. Evaluation of the results for the Kainuu region
In this section we present an evaluation of the employment
effects of Talvivaara mine for the Kainuu region over two periods.
The first evaluation covers the 2009–2014 period for which all
financial statement data are available (see www.talvivaara.com).
The second evaluation comprises a projection for the 2015–2022
period, which is based on full implementation of the plan of corporate restructuring. Our evaluation tool was the dynamic general
equilibrium model, CGE RegFinDyn of the Finnish economy8.
The elements of the scenarios include the actual or assumed
production levels, mining and environmental investments, and the
debt burden. The investments are treated separately from the debt
related issues in the scenarios. Talvivaara had taken out many
loans in order to finance their large mining investments. The environmental accidents forced the company to borrow even more
money in order to finance the necessary environmental rehabilitation measures. The other part of debt is respect the amount
of zinc concentrate that Talvivaara had promised to deliver to
Nyrstar Ltd but had not done so. Nyrstar prepaid for 1.25 million
tons of zinc produced. However, only a fraction of this amount was
ever delivered due to the accidents. The pre-paying by Nyrstar in
effect turned into a considerable loan to Talvivaara, thus Nyrstar
had become one of the biggest creditors when the negotiations
about the corporate restructuring plan were held. Since our model
does not have a financial module, our proxy for the debt burden
was to model it as an extra tax on production, the effect of which
was to increase the costs of production. The extra production tax
that compriced these two parts of the debt burden (specific loans
and undelivered prepaid zinc) worked well in the simulations. We
return to these results in Section 8.1.
We will first concentrate on the employment effects for both
periods. All results are expressed as deviations from the baseline
in person working years. All values and ratios for the results include the direct and multiplier effects. The influence of inflation
was removed from the results.
135
Table 3
The development of the Talvivaara mine over the period 2009–2014.
Years Nickel, tons Zinc, tons Turnover, €
million
Price of nickel, Price of zinc,
$ per ton
$ per ton
2009
2010
2011
2012
2013
2014
Total
14,655
21,809
22,910
17,548
13,684
15,000
–
735
10,382
16,087
12,916
8725
10,469
59,314
3133
25,462
31,815
25,867
13,722
20,816
120,815
7.6
152.2
231.2
142.9
77.6
136.4
747.9
1655
2161
2194
1950
1866
2150
–
turnover, and lead to increased levels of debt. The development of
the mine is presented in Table 3.
The first set annual production target was 30,000 tons of nickel.
This corresponds to about 60,000 tons of zinc annually. Slightly
over half of the target for nickel was reached by the end of 2011.
After the environmental accidents, nickel production fell considerably in 2013. During the negotiations about the corporate
restructuring program in 2014, production increased somewhat.
The falling prices for nickel and zinc have been a challenge for the
mine, and are reflected in the annual turnover figures. Only about
10% of the promised zinc to Nyrstar NV Belgium had been delivered by the end of 2014.
The employment effects for the first period are shown in Fig. 3.
The results suggest that the Talvivaara mine was successful in
creating employment in the first two years of operation, in spite of
the fact that the debt burden had been in effect. The debt burden
eased a little in the third, fourth and sixth years, so that the upturn
in employment seen in 2014 was fostered by the lowering level of
the debt. The employment effects were negative for the fourth and
fifth year after the accidents, when environmental and production
problems began. The results indicate that the mine created a total
of 548 person working years over the 2009–2014 period. This is
the equivalent of a mean annual figure of 91 person working years.
The corresponding influence of the debt burden has been associated with a decrease in employment of 209 person working
years. Since the net overall effect on employment caused by the
mine was positive for the 2009–2014 period, we can conclude that
the mine had a positive economic influence for the Kainuu region
over that period.
8.1. The Years 2009–2014
The Talvivaara mine produced nickel and zinc without mishap
for three years. The environmental accidents caused severe production losses in the years 2012, 2013 and 2014. This reduced the
8
The elasticity values of the model are available upon request.
Fig. 3. The employment effects of the Talvivaara mine activities during the 2009–
2014 period divided between the debt burden and other determinants of
development.
136
H. Törmä et al. / Resources Policy 46 (2015) 127–138
suggest that the Talvivaara mine could create positive employment
for the period 2015–2022. The results indicate that the mine could
create a total of 1207 person working years over the 2015–2022
period. This would be the equivalent of a mean of 151 person
working years annually. The increase in employment could be
largest in the year 2015 when nickel production begins to grow,
and the environment investments are at the highest level. The
debt burden would be considerably lowered for the years 2015 and
2016. The corresponding addition to the employment could be 38
person working years. The effect of the debt burden is expected to
be negative for the years 2017–2020, some 36 person working
years could be lost. There is no effect from the debt burden for the
two last years. The net cumulative effect from the debt burden is
neutral. The results suggest that the downscaling of the debt could
have a positive increase effect on employment for those years for
which the downscaling is the highest.
Fig. 4. The assumed annual production of nickel at the Talvivaara mine projected
for the years 2015–2022.
8.2. The 2015–2022 period
We created a scenario for the future period. The dominating
determinant of growth is the assumption about the path to the
long-run production target of 30,000 tons of nickel. Fig. 4 shows
the assumed path for the scenario. The path starts from an actual
production level, 10,000 tons of nickel for the year 2014. We assumed that nickel production would increase in steady steps toward the target during the full implementation of the rejected
corporate restructuring program that was proposed. Reaching the
target of 30,000 would take eight years.
Further, we assume that the respective price of nickel and zinc
will grow by 9% and 6% according to the long run prediction made
by the World Bank. The downscaling of the mine's debt was to be
between 97% and 99%. This corresponds to the figures suggested in
the rejected corporate restructuring plan. The specifications for
the size of the environmental investments for the years 2015–
2022 are €195 million. This corresponds to official estimates of
required environmental reserves specified by the Centre for Economic Development, Transport and the Environment of Kainuu.
The employment effects for the future scenario are shown in
Fig. 5.
If our assumptions for the scenario are correct, then the results
9. The population forecast
The simulation model in this study, utilized a population
module9 that is explained in Section 7. The main postulation is
that the increase of employment opportunities will also increase
the migration to the Kainuu region where the Talvivaara mine is
located. We modeled in migration and postulate that the regional
unemployment differential compared to the whole country will
determine its development. The corresponding elasticity value
was estimated to be 0.05. Therefore, if the unemployment differential changes in favor of the Kainuu region by one percentage
point, then in migration to Kainuu will increase by 0.05%.
We calculated the population predictions for the future scenario. The results are presented in Fig. 6.
If our assumptions for the scenario are correct, the results indicate that the full implementation of the rejected corporate restructuring plan would have been a tolerable solution for the
Kainuu region. The results also indicate that the mine could not
stop the decrease in population of the region during the period
2015–2020. The population should begin to stabilize because of
the Talvivaara mine activity during the years 2021–2022. The
change of population of the region will be 455 inhabitants. The
corresponding annual mean could be 57 inhabitants. The corresponding numbers for the baseline without the influence from
the Talvivaara mine could be 881 and 110 inhabitants. The
contribution of the mine to the region could be þ426 inhabitants
over the period 2015–2022 compared with the baseline.
10. Summary and conclusions
This study evaluated the employment and population effects of
the Talvivaara mine for the Kainuu region over two periods. The
first period covered the years 2009–2014 for which all economic
data are available. The other evaluation was based on projection of
a scenario for the years 2015–2022. The scenario corresponded to
the time span and actions that were planned for corporate restructuring. Our evaluation tool was the dynamic general equilibrium model, CGE RegFinDyn of the Finnish economy. The elements of the scenarios included the actual or targeted production
levels, mining and environmental investments and the debt burden. The corresponding population forecast was calculated for the
future period.
Our results for the period 2009–2014 indicate that the
Fig. 5. The forecasted employment effects of the Talvivaara mine activities projected for the 2015–2022 period. Note: The full implementation of the draft corporate restructuring program. Assuming a production target 30,000 tons of nickel
at the end of the year 2022.
9
D.Sc Jouko Kinnunen from ÅSUB (www.asub.ax) has designed and programed
the population module.
H. Törmä et al. / Resources Policy 46 (2015) 127–138
Fig. 6. The forecasted total population effects of the Talvivaara mine projected for
the 2015–2022 period.
Talvivaara mine still had a positive cumulative effect on the employment of Kainuu, in spite of the environmental accidents. Since
the net overall effect for employment from the mine was positive
for this period, we can conclude that the mine still has a positive
economic influence for the Kainuu region.
The results also suggest that if the creditors had accepted the
plan for the corporate restructuring for the years 2015–2022 that
was offered, it would have been a tolerable solution for the Kainuu
region. If our assumptions for the scenario are correct, then the
results suggest that the Talvivaara mine could create positive
employment for the period 2015–2022. The results indicate that
the mine could create a total of 1207 person working years over
the 2015–2022 period. This would be the equivalent of a mean of
151 person working years annually. The results also indicate that
the mine would be unable to stop the decrease in population of
Kainuu during the 2015–2020 period. The population would
however be expected to begin to stabilize because of the Talvivaara mine activity during the 2021–2022 period compared with
the baseline.
The authors of this research are fully aware that the situation in
the time span of the activities and investments at the Talvivaara
mine can change very rapidly. It might also be that the extent of
the governmental financing for the mine can also change. Another
important question is how long it will take for the national government to find a new owner for the Talvivaara mine. Considering
these potential developments, we suggest follow-up studies.
The serious environmental accidents at Talvivaara mine lead to
the downfall of the mining company that had a knock-on effect for
the whole of Finland’s mining industry. Finnish legislation emphasizes the principle of sustainability in the use of all natural
resources including the minerals. It is impossible to gain the social
license to operate a mine if the company fails to execute its operations by sustainable methods. The general opinion among experts and ordinary lay persons is that the main causes of the environmental accidents were: the tight timetable, new relatively
untested production technology, and overambitious goals for the
volume of nickel production. It appears that the environmental
accidents at the Talvivaara mine have lifted the issue of possible
environmental threats into agenda. It seems that the monitoring
officials have tightened their grip and are more cautious in setting
the emission limits after the accidents at the Talvivaara mine.
The Kainuu region is small in terms of population, employment
and regional GDP. The Talvivaara mine was expected to give Kainuu a necessary growth stimulus. The investment of over 500
137
Fig. A1. The mechanism between the expected rate of return and capital growth
rate.
Source: Dynamic CGE Modeling Course Material, 2010. Monash University, Centre of
Policy Analysis, Melbourne Australia.
million euros was expected to generate hundreds of new jobs, and
to provide a stable employment for several decades. The big investor had a good story, lots of goals and evidence of progress. It is
possible that even the government officials did not fully grasp the
size and complexity of the environmental risks, and the limits for
the emissions were set too high. No government official has been
brought-to-court, nor blamed for the accident, however.
It is worth mentioning that there are at least two new mines in
Finland that have not had such problems concerning the environment. A nickel and copper mine in Kevitsa that is situated in
Sodankylä North Lapland is a good example. That mine is owned
and operated by international FQM–First Quantum Minerals Ltd. It
invested almost 500 million euros in the mine, and is now producing minerals and making profits. Another good example is the
Suurikuusikko gold mine situated in Kittilä Mountain Lapland and
operated by international Agnico Eagle Finland Ltd. The investment cost was almost 400 million euros. The mine has operated
for a bit longer than Kevitsa, and produces gold profitably. The
population of the municipality of Kittilä has grown by hundreds of
new inhabitants, and has seen a sizeable increase in its tax revenue. Both firms have long experience in mining, and use welltested and established production technologies.
It seems to us that finding a resourceful investor, with long
experience, high-level expertise, careful planning, a reasonable
timetable and an established production technology is the best
way to guarantee success in sustainable mining.
Acknowledgments
This work was supported by the Municipal Council of Kainuu
Finland (Grant code 3999637). Comments received at the Via Futuri 2014 conference in Pécs Hungary are greatly appreciated.
Appendix A
See Fig. A1.
138
H. Törmä et al. / Resources Policy 46 (2015) 127–138
References
Adams, P., Dixon, J., Giesecke, J., Horridge, M., 2010. MMRF: Monash Multi-Regional
Forecasting Model. The Centre of Policy Studies, (CoPS), University of Melbourne Australia, General Paper No. G-223, December 2010.
Bohlmann H., Dixon P., Rimmer M. Van Heerden J., 2014. The impact of the 2014
platinum mining strike in South Africa: an economy-wide analysis ERSA
working paper 478.
Britz, W., 2012. RegCgeEUþ in GAMS, documentation including the Graphical User
Interface. FP7 CAPRI-RD project, WP 3.2 Model development and adaptation –
Regional CGEs, Additional Deliverable: D3.2.4.
Castillo, J., Yu, K., 2014. Liberalizing the Philippine mining industry: a CGE approach
into analyzing its sectoral impacts on the economy. DLSU Bus. Econ. Rev. 23 (2),
48–64.
Centre for International Economics, 2010. The Economic Impact of the Sepon Mine.
Prepared for Lane Xang Minerals Limited, Canberra & Sydney.
Database Industry Online, 2013. Statistical base, Enterprises and industries. Available at: 〈http://www2.toimialaonline.fi/〉.
Downes, P., Hanslow, K., Tulip, P., 2014. The Effect of the Mining Boom on the
Australian Economy. Reserve Bank of Australia, Sydney, Research Discussion
Paper.
Dynamic CGE modeling course material, 2010. Monash University, Centre of Policy
Analysis, Melbourne Australia.
Ejdemo, T., Söderholm, P., 2011. Mining investment and regional development: a
scenario-based assessment for Northern Sweden. Resour. Policy 36, 14–21.
Geological survey of Finland, 2009. Main metallogenic zones and metallic ore
mines of Finland.
Geological Survey of Finland, 2010. Finland’s Minerals Strategy. Available at: 〈http://
projects.gtk.fi/export/sites/projects/minerals_strategy/documents/FinlandsMi
neralsStrategy_2.pdf〉.
Honkatukia, J., Ahokas, J., 2012. Suomen talouden rakenteellinen kehitys finanssikriisin jälkeen – Kysyntä- ja tuotantorakenteiden muutos vuosina 2012–
2030. VATT Tutk. 169.
Kokko, M., 2014. The Mining Industry. Ministry of Employment and the Economy.
Available at: 〈http://www.temtoimialapalvelu.fi/files/2253/Kaivosteollisuus_
marraskuu_2014.pdf〉.
Laukkonen, J., Törmä, H., 2014. Finnish Mining Sector's Impacts and Challenges
Between Years 2010 and 2020. University of Helsinki, Ruralia Institute, Reports
136, Finland.
LSE Enterprise, 2011. Study on the Impact of the Single Market on Cohesion: Implications for Cohesion Policy, Growth and Competitiveness. European Commission, Directorate-General Regional Policy, Directorate Policy Development.
Peach, J., Popp, A., 2008. The Economic Impact of Proposed Uranium Mining and
Milling Operations in the State of New Mexico. Office of Policy Analysis, Arrowhead Center, New Mexico State University.
Rolfe, J., Akbar, D., Ivanova, G., Greer, L., Donarthy, P., Timmer, V., Yabsley, B., 2011.
Social and economic impact assessment of the Codrilla coal mine project.
Centre for Environmental Management, CGUniversity, Australia, Prepared for
McCollum Environmental Management Services Pty Ltd.
Rutherford, T., Törmä, H., 2010. Efficiency of Fiscal measures in preventing out
migration from North-Finland. Reg. Stud. 44 (4), 465–475.
Skurla, J., Jacobson, J., Kasim, T., Solem, L., Almquist-Minko, V., 2009. The Economic
Impact of Ferrous and Non‐Ferrous Mining On the State of Minnesota And on
the Arrowhead Region and Douglas County, WI. University of Minnesota, Bureau of Business and Economic Research, Research Report.
Statistics Finland, 2014. Regional Accounts. Available at: 〈http://www.stat.fi/index_
en.html〉.
Törmä, H., 2008. Do small towns development projects matter, and can CGE help? J.
Spat. Econ. Anal. 3 (2), 247–268.
Törmä, H., Lehtonen, H., 2010. Macroeconomic and welfare effects of the CAP reform and further decoupling of agricultural support in Finland: a CGE modelling approach. Food Econ. – Acta Agric. Scand. Sect. C 6, 73–87.
Törmä, H., Reini, K., 2009. The Effects of the Transport Infrastructure Investments
and the Operation Phases of Pajala–Kolari and Sokli Mining Projects to the
Accumulation of State Taxes and Tax-like Charges. University of Helsinki, Ruralia Institute, Reports 42, Finland.
Törmä, H., Zawalinska, K., Blanco-Fonseca M., Ferrari E., Jansson T., 2010. Regional
CGE model layout with a focus on integration with the partial equilibrium
models and modelling of RD measures. FP7 CAPRI-RD project, Deliverable 3.2.1.
Törmä, H., Zawalinska, K., 2010. Methodological description of the CGERegEU. FP7
CAPRI-RD project, Deliverable 3.2.2.
Törmä, H., Zawalinska, K., 2011. Final documentation of the CGERegEU þ model.
CAPPRI-RD Deliverable 3.2.3.
Weisskoff, R., 2010. Economic Impact of Mining on Levy County, Florida: A Strategic
View.
Wittwer, G., Horridge, M., 2010. Bringing regional detail to a CGE model using
census data. Spat. Econ. Anal. 5 (2), 229–255.
Ye, Q., 2008. Commodity booms and their impacts on the Western Australian
economy: the iron ore case. Resour. Policy 33 (2), 83–101.
Zawalinska, K., 2009. Instruments and Effects of European Union Support for Regional Rural Development in Poland. Institute of Poland Agricultural Development, Polish Academy of Sciences, Warsaw.