MTT is publishing its research fi ndings in two series of publications: MTT Science and MTT Growth. The MTT Science series includes scientifi c presentations and abstracts from conferences arranged by MTT Agrifood Research Finland. Doctoral dissertations by MTT research scientists will also be published in this series. The topics range from agricultural and food research to environmental research in the fi eld of agriculture. MTT, FI-31600 Jokioinen, Finland. Tel. +358 3 4188 2327, email julkaisut@mtt.fi 4 Land leasing, land degradation and agricultural productivity in Finland Doctoral Dissertation Sami Myyrä MTT CREATES VITALITY THROUGH SCIENCE www.mtt.fi /julkaisut 4 Land leasing, land degradation and agricultural productivity in Finland Doctoral Dissertation Sami Myyrä Academic Dissertation: To be presented, with the permission of the Faculty of Agriculture and Forestry of the University of Helsinki, for public criticism in Lecture Room 13, Fabianinkatu 33 (3. floor), on November 20th 2009, at 12 o’clock. 2 MTT SCIENCE 4 ISBN 978-952-487-246-1 (Print) ISBN 978-952-487-247-8 (Electronic) ISSN 1798-1824 (Printed version) ISSN 1798-1840 (Electronic version) http://www.mtt.fi/mtttiede/pdf/mtttiede4.pdf Copyright MTT Agrifood Research Finland Sami Myyrä Distribution and sale MTT Agrifood Research Finland, Media and Information services, FI-31600 Jokioinen, phone +358 3 41881, e-mail julkaisut@mtt.fi Printing year 2009 Cover picture Riitta Myyrä Printing house Tampereen Yliopistopaino Juvenes Print Oy Supervisors: Professor Kyösti Pietola Economic Research MTT Agrifood Research Finland Professor Matti Ylätalo Department of Economics and Manage- ment, University of Helsinki, Finland Pre-reviewers: Professor Jari Kuuluvainen Department of Forest Economics University of Helsinki, Finland Associate Professor Jack Peerlings Department of Social Sciences Agricultural Economics and Rural Policy Group, Wageningen University, e Netherlands Opponent: Professor Pekka Ilmakunnas Department of Economics Helsinki School of Economics, Finland Custos: Professor Matti Ylätalo Department of Economics and Manage- ment, University of Helsinki, Finland MTT SCIENCE 4 3 Land leasing, land degradation and agricultural productivity in Finland Sami Myyrä MTT Agrifood Research Finland, Economic Research, Luutnantintie 13, 00410 Helsinki, firstname.lastname@mtt.fi fixed-term lease contract. e results sug- gest that as the probability of non-renew- al of the lease contract increases, farmers quickly reduce investments in irreversible land improvements and, thereafter, yields gradually decline. e simulations high- lighted the observed trends of a decline in land improvements on land parcels that are cultivated under lease contracts. Land tenure has resulted in the neglect of land improvement in Finland. is study aimed to analyze whether these challenges could be resolved by a tax policy that en- courages land sales. Using Finnish data, real estate tax and a temporal relaxation on the taxation of capital gains showed some potential for the restructuring of land own- ership. Potential sellers who could not be revealed by traditional logit models were identified with the latent class approach. ose landowners with an intention to sell even without a policy change were sensi- tive to temporal relaxation in the taxation of capital gains. In the long term, productivity and especial- ly productivity growth are necessary condi- tions for the survival of farms and the food industry in Finland. Technical progress was found to drive the increase in productivity. e scale had only a moderate effect and for the whole study period (1976–2006) the effect was close to zero. Total factor Abstract This study analysed whether the land tenure insecurity problem has led to a decline in long-term land im- provements (liming and phosphorus ferti- lization) under the Common Agricultural Policy (CAP) and Nordic production con- ditions in European Union (EU) coun- tries such as Finland. e results suggests that under traditional cash lease contracts, which are encouraged by the existing land leasing regulations and agricultural subsidy programs, the land tenure insecurity prob- lem on leased land reduces land improve- ments that have a long pay-back period. In particular, soil pH was found to be sig- nificantly lower on land cultivated under a lease contract compared to land owned by the farmers themselves. e results also indicate that land improvements could not be reversed by land markets, because land owners would otherwise have carried out land improvements even if not farming by themselves. To reveal the causality between land ten- ure and land improvements, the dynam- ic optimisation problem was solved by a stochastic dynamic programming routine with known parameters for one-period re- turns and transition equations. e model parameters represented Finnish soil qual- ity and production conditions. e deci- sion rules were solved for alternative likeli- hood scenarios over the continuation of the 4 MTT SCIENCE 4 productivity (TFP) increased, depending on the model, by 0.6–1.7% per year. e results demonstrated that the increase in productivity was hindered by the policy changes introduced in 1995. It is also evi- denced that the increase in land leasing is connected to these policy changes. Land institutions and land tenure ques- tions are essential in agricultural and rural policies on all levels, from local to inter- national. Land ownership and land titles are commonly tied to fundamental politi- cal, economic and social questions. A fair resolution calls for innovative and new so- lutions both on national and internation- al levels. However, this seems to be a prob- lem when considering the application of EU regulations to member states inher- iting divergent landownership structures and farming cultures. e contribution of this study is in describing the consequenc- es of fitting EU agricultural policy to Finn- ish agricultural land tenure conditions and heritage. Key words: land tenure, insecurity, productivity, taxation MTT SCIENCE 4 5 Pellon vuokraus, perusparannusten laiminlyönnit ja maatalouden tuottavuuskehitys Suomessa Sami Myyrä MTT (Maa- ja elintarviketalouden tutkimuskeskus), Taloustutkimus, Luutnantintie 13, 00410 Helsinki, etunimi.sukunimi@mtt.fi Tiivistelmä Tässä tutkimuksessa tarkastellaan pel-lon vuokrauksessa esiintyvän pel-lon hallintaoikeuden epävarmuu- den vaikutusta pellon perusparannuksiin. Tutkimuksen erityispiirteenä on suomalai- sen kehittyneen teollisuusmaan, maatalou- den kannalta pohjoisen sijainnin ja Euroo- pan Unionin yhteisen maatalouspolitiikan kombinaatio. Tulokset osoittavat, että tu- kijärjestelmän sekä pellonvuokralainsää- dännön suosimat pellonvuokrauksen kä- teismarkkinat johtavat pitkävaikutteisten ja peruuttamattomien perusparannusten laiminlyönteihin. Ilmeistä on myös se, että pellonvuokramarkkinat eivät toimi. Toimivilla pellonvuokramarkkinoilla pel- lon omistajankin olisi kannattanut tehdä perusparannukset, vaikka he eivät viljeli- sikään peltojaan itse. Suomessa eri tyisesti vuokrapeltojen pH-taso on tilastol lisesti merkitsevästi alem pi kuin omistajan itse viljelemien peltojen vastaava. Pellon vuokrauksen ja perusparannus- ten kausaalisuhteen selvittämiseksi vuok- ra viljelijän yli ajan ulottuvat kal kitus- ja fosforilannoituspäätökset ratkais tiin dy- naamisen optimoinnin menetelmillä. Dy- naamisessa optimoinnissa hyödynnettiin aikaisempaa tutkimustietoa satorespons- seista sekä maan helppoliukoisen fosforin ja pH:n kehi tyksestä viljelymaassa. Kal- kitus- ja fosforilannoitussäännöt ratkais- tiin kiinteille, viiden vuo den vuokraso- pimuksille simuloimalla eri skenaarioita vuokrasopimuksen uusiutumistodennä- köisyydestä. Vuokraviljelijän ennakoima vuokrasopimuksen uusiutumi sen toden- näköisyys osoittautui keskeiseksi muuttu- jaksi kalkitus- ja fosforilannoituspäätök- sissä. Todennäköisyyden laskiessa myös kalkitus ja fosforilannoitus laskevat nope- asti, mikä kääntää pellon tuottaman sa- don laskuun. Tulokset tuovat selvästi esiin pellonvuokraukseen liittyvän pellon hal- lintaoikeuden epävarmuuden ja peruspa- rannusten välisen kausaaliyhteyden ja vah- vistavat pellonvuokramarkkinoilta teh tyjä havaintoja. Pitkällä aikavälillä pellon hallintaoikeuden epävarmuudesta aiheutuvat perusparannus- ten laiminlyönnit saattavat johtaa osan pel- loista matalan tuottavuuden loukkuun, jos- sa ei vuokraviljelijällä eikä pellon omistajalla ole taloudellisia kannusteita pellon perus- parannuksiin. Verotuksen kehittämistä tar- kasteltiin eräänä vaihtoehtona näiden on- gelmien ratkaisemiseksi. Verotuksellisina keinoina tutkimuksessa olivat mukana sekä pellon kiinteistövero että pellon myynti- voittoveron väliaikainen huojennus, joiden 6 MTT SCIENCE 4 molempien arvioitiin vaikuttavan maan- omistajien päätöksiin myydä peltonsa. Pe- rinteisillä logit-malleilla potentiaalisia pel- lon myyjiä ei pystytty kytkemään vahvasti mihinkään pellonomistajia kuvaavista so- siodemografisista taustamuuttujista. Tosin aikaisemmat pellon myyntikokemukset näyttivät olevan yhteydessä pellon myyn- tiaikomuksiin. Latenteista pellonomista- jaryhmistä saatiin lisätietoa latent class -menetelmällä. Havaintojen perusteella ne pellonomistajat, joilla on myyntiaiko- muksia, tarttuvat herkästi väliaikaisen pel- lon myyntivoittoveron huojennuksen kal- taisiin pellon myyntikannusteisiin. Maatalouden myönteinen tuottavuuske- hitys edistää Suomen elintarviketuotan- non kilpailukykyä. Viljanviljelyn tuotta- vuuskehitys on ollut kuitenkin hidasta, ja se on edelleen hidastunut EU-jäsenyyden ja lisääntyneen pellonvuokrauksen myö- tä. Pitkällä aikavälillä viljatilojen tuotta- vuuskehitys on ollut 0,6–1,7 % vuodessa. Havaittu tuottavuuskehitys on aiheutunut lähes yksinomaan teknologisesta kehityk- sestä. Tuotannon laajuudesta aiheutuvilla mittakaavaeduilla on ollut vain pieni vai- kutus tuottavuuskehitykseen. Pellon hallintasuhteilla on keskeinen ins- titutionaalinen asema maatalous- ja maa- seutupolitiikassa niin kansallisella kuin kansainväliselläkin tasolla. Pellon omis- tus- ja käyttöoikeudet liittyvät perustavaa laatua oleviin poliittisiin, taloudellisiin ja sosiaalisiin kysymyksiin. Tulevaisuudessa oikeudenmukaisten ratkaisujen kehittämi- nen vaatii uusia innovaatioita. Näyttää kui- tenkin siltä, että uusien kansainvälisten so- pimusten ja säännöstöjen, kuten Euroopan Unionin yhteisen maatalouspolitiikan, so- veltaminen laajenevassa unionissa on jäsen- maittain pitkälle eriytyneistä peltomarkki- noista johtuen entistä haastavampaa. Tämä tutkimus pyrkii osaltaan myötävaikutta- maan pellonhallinta oikeutta koskevien ins- titutionaalisten kysymysten ratkaisemises- sa ja yleisemminkin osallistuu keskusteluun yhteisen maata louspolitiikan soveltamisesta suomalaisille pellonvuokramarkkinoille. Avainsanat: pellon vuokraus, epävarmuus, perusparannukset, tuottavuuskehitys, verotus MTT SCIENCE 4 7 I thank Professor Frank Scrimgeour and all the Kiwi people, who generously opened their friendship for me and my family. I thank the reviewers of this thesis, Pro- fessor Jari Kuuluvainen and Professor Jack Peerlings, the language consultant Dr Roy Siddall and the research assistant Jaana Ahlstedt for their valuable comments and practical help in all stages of preparing the articles and/or the summary part included in this thesis. In addition to the generous funding from MAKERA, Salaojituksen tutkimusyhdistys ry and Huoltovarmuuskeskus, my studies and this thesis were financed by the Kyösti Haataja Foundation, the Markku Nevala Foundation, the August and Aino Tiura Foundation, University of Helsinki Funds and the Finnish Cultural Foundation. I acknowledge, with deepest gratitude, all these organizations. I am a happy man. I have shared my life with a woman who makes my life com- plete. Without your encouragement, ded- ication, warmness and love I would not have gone this far in my studies. ank you Riitta! Hannes and Aino, your un- questioned joy of life is something that I should have learned better. Please keep on teaching me! The research included in this thesis was carried out in the Economics unit of Agrifood Research Finland (MTT). In all my studies I have enjoyed the multidisciplinary cooperation and team work of people passionately involved in re- search. It has never been a question of or- ganizational forms, but the dedication and passion of people who have carried out the research. I thank all the people work- ing in the Economics unit for their skilful and generous help in numerous tasks. It has also been a pleasure to take part in ac- ademic as well as “coffeedemic” conversa- tions, where we have reformed the agricul- tural policy so many times. I am grateful to my teacher, supervisor and co-author, Professor Kyösti Pietola, for his extensive scientific input as well as his administrative and practical help in all the work included in this thesis. With- out his endless support, this thesis would not have achieved its current form. I also thank my other teachers and co-authors, Professor Matti Ylätalo, Dr Eija Pouta, Professor Markku Yli-Halla, Biometri- cian Elise Ketoja, Dr Timo Sipiläinen and Pekka Pihamaa, MSc, for their valuable contributions. During my studies I had a great opportunity to visit the University of Waikato Management School in Ham- ilton, New Zealand. For this opportunity Acknowledgements 8 MTT SCIENCE 4 is thesis consists of the Summary and the following four articles. I Myyrä, S., Ketoja, E., Yli-Halla, M. & Pietola, K. 2005. Land improvements under land tenure insecurity : the case of pH and phosphate in Finland. Land Economics 81(4): 557– 569. II Myyrä, S., Pietola, K. & Yli-Halla, M. 2007. Exploring long-term land improvements un- der land tenure insecurity. Agricultural Systems 92: 63–75. III Myyrä, S. & Pouta, E. Farmland owners’ land sale preferences: can they be affected by taxation programmes? Forthcoming in Land Economics. IV Myyrä, S., Pihamaa, P. & Sipiläinen, T. Productivity growth on Finnish grain farms from 1976–2006: a parametric approach. Forthcoming in Agricultural and Food Science. Reprints of the original articles are published with the permission of the respective copyright owners. Myyrä, S., E. Ketoja, M. Yli-Halla & K. Pietola ”Land improvements Under Land Tenure Inse- curity: The Case of pH and Phosphate in Finland. Originally published in Land Economics 81.4 (2005): 557–569. 2005 by the Board of Regents of the University of Wisconsin Sys- tem. Reproduced Courtesy of the University of Wisconsin Press. Myyrä, S. & E. Pouta. “Farmland Owners’ Land Sale Preferences: Can They be Affected by Taxation? Originally published in Land Economics 86.2 (2009). 2009 by the Board of Re- gents of the University of Wisconsin System. Reproduced Courtesy of the University of Wis- consin Press. Authors’ contribution Articles I and II are based on a multidisciplinary research project including soil science, bi- ometrics and economics. The development of the theoretical framework, the empirical ap- proach and the data collection were mainly performed by Mr Sami Myyrä. Statistical meth- ods were adapted and the actual model fitting was carried out by Ms Elise Ketoja. All authors contributed to the writing of the report, the first author contributing most and being most responsible for the text. Article II is a pure numerical application performed under the framework of dynamic pro- gramming (DP). DP was introduced to the Finnish agricultural economists’ toolbox by Dr Kyösti Pietola. This particular application was, nevertheless, designed, programmed and calibrated to the data by Mr Sami Myyrä with the guidance of Dr Pietola and Dr Yli-Hal- la. All authors contributed to the writing of the report, but Myyrä was responsible for the manuscript. The first draft of Article III was written by Mr Sami Myyrä during his visit to the University of Waikato in New Zealand. Latent class methodology was introduced by Mr Sami Myyrä and further extended in collaboration with Dr Eija Pouta. Both authors jointly contributed to the final text. The empirical approach and the data collection for Article IV were mainly performed by Mr Sami Myyrä and Mr Pekka Pihamaa. The model fitting and drawing of conclusions were mostly carried out by Mr Sami Myyrä. The theoretical foundation for the paper was clarified in collaboration with Dr Timo Sipiläinen. All authors contributed to the writing of the report. List of original publications MTT SCIENCE 4 9 Contents 1 Introduction ...................................................................................................10 1.1 Background and literature ........................................................................... 10 1.2 Framework of the study .............................................................................. 13 1.3 Objectives of the study ............................................................................... 16 1.4 Outline of the thesis ................................................................................... 17 2 Results and discussion ....................................................................................18 2.1 Land tenure insecurity in Finland ............................................................... 18 2.2 Lease holder’s profit maximization and land tenure insecurity ..................... 20 2.3 Taxation and land sale ................................................................................. 23 2.4 Long-term productivity changes in grain production .................................. 25 3 Conclusions ....................................................................................................27 4 Yhteenveto (summary in Finnish) ...................................................................29 References ............................................................................................................32 Appendices...........................................................................................................35 10 MTT SCIENCE 4 Background 1.1 and literature Agriculture is a long-term operation. A land- owner has a planning horizon over genera- tions when deciding on land improvements, but for a tenant the planning horizon may be short. It has therefore often been argued that tenants tend to overexploit the land. e question is how land tenure contracts should be formulated to avoid overexploi- tation. Also, are the agricultural markets ef- ficient enough to transform incentives for a tenant to make land improvements that are necessary from the society’s perspective? Gradual land degradation may not imply an immediate or sudden loss of agricultural productivity. Moreover, technical progress, an increased scale and other improvements might have outweighed slowly proceeding degradation processes in agricultural lands. In addition to land tenure insecurity, land degradation might result from other causes. Poverty, population pressure, a lack of effi- cient markets for commodities, inputs, and externalities have been mentioned as factors causing land degradation, among others. Al- though the reasons for degradation vary sig- nificantly over international market regions, they may all be key factors behind the result- ing low productivity trap, depending on the local land ownership structure, initial land rights as well as social structures and norms. e highly mechanised, market-orientated and developed conditions in Finnish agri- culture are not typical for a low productivity trap in cases earlier described in agricultural economic literature. erefore, the follow- ing sections describe in more detail the back- ground and concepts in Finnish agricultural conditions, as a part of common agricultural policy (CAP) in European Union (EU). e soils of Finland have been formed from acidic rock and pH values in agricultural soils in the country are commonly below the recommended level. erefore, liming is one of the basic land improvements used to maintain good yields. Good yields also en- sure an efficient harvest of nutrients and this way prevent their runoff, which is extreme- ly difficult to capture later as nutrients are transported to watersheds. From the 1960s until the 1990s there was a slight but stead- ily increasing trend in soil pH, but particu- larly during the last decade liming has been practiced far less than recommended (Vilja- vuuspalvelu 2009, Yearbook of Farm Statis- tics). In order to maintain soil productivity, soil pH should be raised by liming. Land improvements have generally declined and the demand for lime, for example, de- creased by 49% between 1994 and 2004, even though the total arable land area slight- ly increased during this period. ere are signals that the greatest declines in land im- provements have been in land parcels that have been cultivated under lease contracts. Land improvements and land leasing in Fin- land have followed opposite trends, which indicates a potential land tenure insecurity problem (Yearbook of Farm Statistics). Sustainable land management practises are most commonly constrained by lack of secure access to private property. e unrestricted access to commonly owned land recourses has also been noted, but this is a different phe- nomenon referred to as the “tragedy of the commons” (Hardin 1968). Such a problem does not currently exist in Finnish agricul- ture, because more than 97% of agricultural lands in Finland are privately owned. Introduction1 MTT SCIENCE 4 11 In Finland, the share of agricultural land under lease contracts has doubled within the last 15 years (Figure 1). e standard land lease contract is a short-term contract with a fixed duration and a fixed cash lease payment per year and per unit of leased area. About 40% of all lease contracts have a duration of five years. With only a few exceptions, the annual cash lease payment is fixed per hectare of land when the con- tract is signed (Myyrä 2004). Under this contract the farmer supplies the optimal amount of inputs and has free access to the land’s unprized attributes, i.e. productivity stock. Farmers do not face the full cost of usage of these attributes under fixed cash lease contracts (Allen and Lueck 2002). However, they own the entire crop. e so- cially optimal level of land improvements refers to a level that is reached at a com- petitive market with perfect information and perfectly defined property rights (Mas- Colell et al. 1995, pp. 325–328). e so- cial optimum gives the maximum welfare for society as a whole, including landown- ers and tenants. It is noteworthy that in well-developed rich countries, land tenure insecurity has re- ceived little attention in empirical stud- ies and the economic literature1 (Allen & Lueck 2002). Although land tenure inse- curity is not yet a dominating problem in these countries, it may become such, par- ticularly in Less Favoured Areas2 (LFA), where production costs are high and yields are low. e trend towards a more liberal- ised food market and internationally har- monised agricultural policies has reduced the marginal value products (MVP) for ag- ricultural inputs in LFA areas, where the possibilities of farmers to adjust to these trends are limited. Figure 1. Development of land tenure in Finland (Yearbook of Farm Statistics). 1 Databases searched: Agricola, EconLit, Web of Sci- ence and Green file. Publications since 2007. 2 Less favoured areas (LFA): ese are areas, such as mountainous and hilly areas, within the Euro- pean Union where farming is relatively difficult. In these areas farmers may receive compensatory al- lowances within the context of measures that are fi- nanced by the rural development policy (Axis 2). Scource: http://ec.europa.eu/agriculture/glossary/ index_en.htm#lfa Accessed 7.9.2009 were searched. Accessed 5.3.2009 and 7.9.2009. Proportion of agricultural land under lease contracts in Finland 0 5 10 15 20 25 30 35 40 19 74 19 8 0 19 8 7 19 8 8 19 8 9 19 9 0 19 9 1 19 9 2 19 9 3 19 9 4 19 9 5 19 9 6 19 9 7 19 9 8 19 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 % 12 MTT SCIENCE 4 Furthermore, the tendency to decouple subsidies from production has been shown to increase the willingness of farmers to take risks in production (Kondouri et al. 2009). e risk-reducing aspect of land improvements is weakening because the connection between agricultural produc- tivity and profitability is faded out by de- coupled subsidies. Some fundamental land improvements such as drainage systems and developments in farm structure might be ignored as farmers become willing to take greater production risks. Because the pay-back periods of irreversible land im- provements are longer, we may expect land improvements in the less favoured areas of northern Europe to decrease below the so- cially optimal levels if farmers are confront- ed with significant land tenure insecurity. One exception in the current trend of agri- cultural economics literature on land ten- ure insecurity is the recognition of the soil erosion problem in the United States and Canada (Lee 1980, Soule et al. 2000, Fra- ser 2004). However, it has been a challenge to find empirical measures to connect land tenure and land degradation or best man- agement practices (Prokopy et al. 2008). Like most empirical research, Soule et al. 2000 examined the adoption of conser- vation practices. ey reported a decreas- ing influence of cash leasing over time, but found no connection between the top soil depth and the tenure system. is is impor- tant, because a cash lease contract might appear because of land quality and induce a self-selection bias (Allen and Lueck 2002, Huffman and Fukunag 2007). Lee (1980) compared the real rate of soil erosion be- tween different tenure groups but observed no significant connection between land tenure insecurity and land degradation, even though the dataset contained 70 000 primary sampling units. Most of the studies have used the number of land improvements, crop rotation, land improvement costs and other meas- ures as proxies of ongoing developments in land fertility under different land ten- ure regimes. It is striking that, the differ- ences in local soils and other conditions make it difficult to assess whether predict- ed causal connections between land man- agement and land degradation / fertility are valid. For example, in Finland, pota- toes and other intensively grown annu- als are planted in owner-operated fields, whereas grassland, and fallow areas, oc- cur in leased plots. However, this does not necessarily mean overexploitation of farm- ers’ own fields and land improvements in leased plots, as it does, for example, in Brit- ish Columbia (Fraser 2004). From a statistical point of view, causal con- nections can be validly examined only by using corresponding control variables as indicators. e difficulty in Finnish data is that pure lease farming or lease farm- ers do not exist. Farmers always have some land of their own and the division of land improvements between owned and leased plots could not be tracked. Another typi- cal feature is that some land improvements in Finland, such as liming, are carried out only once every five to ten years, thus cau- sality between land tenure and land im- provements is not always clear. In the early stages of this study it already became appar- ent that there are no valid data to indicate appearance of the land tenure insecurity in Finland based primarily on corresponding control variables for land degradation, but the lacking causality could be modelled and calibrated for Finnish conditions. It is known that simple fixed duration and fixed cash lease payment contracts are not the most efficient in internalizing the ex- ternals caused by land degradation (Soule et al. 2000, Huffman and Fukunag 2007). However, they have remained persistent and popular in Finland for several reasons. First, fundamental co-operation, like share- cropping and share milking between a ten- ant and a landowner, is not recognised in the Common Agricultural Policy (CAP) of the European Union (EU) (Swinnen et al. MTT SCIENCE 4 13 2009). Second, land leasing has been di- rectly steered by a variety of policy regu- lations in Europe. Particularly in Finland, these regulations still favour short dura- tion lease contracts. ird, landowners feel a sense of insecurity over the subsidy en- titlements of decoupled subsidies. is is an essential and driving force towards lease contracts limited to the growing season only. Without the capitalisation of subsi- dies in land, land prices will significantly drop (Pyykkönen 2006). Fourth, cash lease contracts have low information require- ments and may therefore result in a reason- able short-term outcome even if the market is informationally inefficient in the long term. In Finland, more than 40 000 people manage agricultural land acquired through inheritance, which accounts for one-tenth of the entire field area. ose inheriting land are not necessarily familiar with agri- culture, and especially where siblings joint- ly inheriting land have to reach a united decision over its usage, they may aim for practical, but not long-term land manage- ment planning. Cash lease contracts are well suited to people whose aim is to trans- form fixed capital into cash. Fifth, the col- lateral value of agricultural land in Finland does not count for the land improvements i.e. in some cases land improvements could be reduced without reducing the collat- eral value of that land and increasing in the price of capital. Financial markets cur- rently lack knowledge of the importance of land productivity to land prices (Ryan et al. 2001). Framework of the study1.2 A downward spiral including land ten- ure, disinvestments in land and decreas- ing or hindered productivity in agricul- ture also exist in developed countries. is situation is known as the low productivi- ty trap3 (Nkonya et al. 2008) (Figure 2). While the agricultural economics litera- ture in well developed countries has not dealt with the low productivity trap, the literature on this problem from develop- ing countries might provide suitable con- ceptualisation. e terms “welfare poverty” and “investment poor” are also appropri- ate in highly mechanized and well-educat- ed production environments. is is due to the lack of economic incentives, com- plicated land ownership and leasing insti- tutions and policy disincentives. Reardon and Vosti (1995) defined investment poor in the context of agricultural land invest- ments as “a situation where farmers are not able to make minimum investments in the resource base to enhance the sustainable quality of agricultural land or are not able to reverse resource degradation.” Pover- ty is usually associated with disability, but this is unlike the case in Finnish agricul- ture. is calls for a study of other reasons explaining the current trends in Finland. I propose that investment poorness is caused by a lack of economic incentives. Follow- ing Reardon and Vosti (1995), it could be argued that a household may choose to use the agricultural surplus for consumption, savings, or investments of other types than in the land due to external conditioning factors. When combined with substantial land tenure insecurity, investments else- where than in land improvement appear more attractive. Undeveloped markets and a lack of incen- tives are comparable because they might hinder investments in land or the abili- ty of households to convert one form of wealth into another. In Finnish cash lease contracts, land improvements are trans- formed into cash, because markets do not 3 Certain other economic terms or processes such as the “lock-in effect” or “race to the bottom” could also be applied and examined. However, these are respectively related to taxation and competition be- tween nations or states. e term “low productivity trap” is also associated with labour economics, but has continuously been used in the case of land ten- ure insecurity. e lock-in effect is referred to lat- er when pathways out of the low-productivity trap are considered. 14 MTT SCIENCE 4 provide long term incentives for land im- provements. e penalising feedback from not investing in land improvement is also lacking. Because of these factors, agricul- ture is likely to face serious problems. Mar- kets are not efficient enough to steer land asset prices to be determined by income- generating factors such as land productivity (Myyrä and Scrimgeour 2008). How ever, the debate over the connection between land productivity and land prices on ag- ricultural lands is still open and shown to be very complicated (Gutierrez et al. 2007, Kataja 2008). Incentives are not only limited in land sales, but also on rental markets. is is a largely accepted land tenure insecurity problem, which limits farm household net returns from land improvement investment by making the future income stream un- certain. A lower discount rate and a longer planning horizon are the key factors to en- courage conservation decisions by increas- ing the present value of expected net reve- nues and by allowing sufficient time to pay back conservation investments. e contribution of land leasing to poor land management and land degradation is implicitly assumed, but not often em- pirically illustrated, at least in rich devel- oped countries. It is argued that well-de- veloped institutional arrangements and efficient asset markets should solve prob- lems caused by land tenure insecurity (Mc- Connell 1983). Hence, economic literature analysing the implications of land tenure insecurity has for the most part focused on developing countries, where asset markets are likely to exhibit inefficiencies (Sjaastad and Bromley 1997, Li et al. 1998, Deinin- ger 2003, Deininger and Chamorro 2004). In these countries, land tenure insecurity has significant implications, not only for land improvements but also for the wel- fare of households and society. Although the problems caused by land tenure inse- curity are apparent, empirical studies have not been able to clearly identify and char- acterise their effects. e data have not been informative enough, because household ac- cess to farming inputs and markets in de- veloping countries can be severely restrict- ed by institutional, financial and economic Figure 2. Low productivity trap. (Adapted from Nkonya et al. 2008). 8???% ??????? 9????????? ????????????0 ??????????????0 ????????0%?????? 8???% ??????????? 7????????? ????????????% ??%??????????? 8??? ???? 7????????% ????????????% ???????????? 3 6 4 5 ?????1????%???? ??????????%2% ????%?????? 9???????%???%???????????? 1%?????????%????????? 1%????????%????????% %%%?????? 1%???????? MTT SCIENCE 4 15 factors, as well as by land tenure insecuri- ty. us, it has been difficult to isolate and identify the impact of land insecurity in sta- tistical analysis (Holden and Hailu 2002, Lichtemberg 2007, Holden et al. 2009). e links between land leasing and land degradation (links 1 and 2 in Figure 2) have also been questioned (Sjaasted and Bromley 1997, Allen and Lueck 2002, Lichtenberg 2007). e key point raised is that rental status is not an exogenous but an endogenous variable and the actions of landowners have been underestimated. In Finnish conditions this critique is not as valid as it is in the U.S., because the contract terms are limited by the govern- ment and the form of share cropping is un- known; thus, the contract terms are exoge- nous in Finland. In addition, the number of landowners exceeds the number of ten- ants by almost ten-fold, and furthermore, the competition between tenants is limit- ed due to the sparsely populated rural ar- eas. Rental prices of €0 /ha are not rare. e average landowner in Finland rent- ing out farmland owns 3.35 ha, while his/ her counterpart in the U.S. owns 109.35 ha (USDA, Myyrä et al. 2008). How ever, liming, the most important land improve- ment, is easily verifiable and the endog- enous nature of land tenure contract ter- mination could bee introduced. Finnish rental market conditions may imply the combination of maximizing the net val- ue of production under the lease period, but the expected present value of land will only be partially achieved. e research problems are related to the levels of con- trol variables deterring land productivi- ty and the following stock levels of land improvements achieved when tenure con- tracts are short and the renewal of the con- tracts is uncertain and exogenously given. ese questions were addressed in two ar- ticles. e tenant’s problem of maximiz- ing the net value of barley production in Finnish conditions is described in Article II and evidence of the land tenure insecu- rity problem based on the empirical foun- dation of stock variables of land improve- ments is presented in Article I. It is a challenging task to statistically dem- onstrate the impact of land management on the productivity of farmland at the ag- gregate level (link 3 in Figure 2). e ef- fects of certain land improvements at the field plot level have been studied more fre- quently under Finnish farming conditions (Kemppainen et al. 1993, Saarela et al. 1995, Turtola and Ylivainio 2009). Doc- umented effects can be used when opti- mal land management by tenants is pinned out (Article II). Nevertheless, on the aggre- gate level we do not know the actual differ- ences in land management between leased and farmer-owned plots. A particular fea- ture in Finnish agriculture is that farmers only lease additional land to expand the size of their farming operations. Because most of the accounts of farms are carried out at the aggregate farm level, we could not observe how farm inputs are allocated between farmer-owned and leased land. However, certain statistics describing land quality and fertility can be separately ob- served for each plot of the farm, and these data can be used to identify differences be- tween owned and leased plots to justify the land tenure insecurity problem. is is a sum of links 1, 2, and 3 in figure 2. To obtain aggregate-level measures of the effect of land tenure on farmland produc- tivity, the productivity of Finnish grain farms between 1976 and 2006 have to be examined. Productivity changes could thereafter be compared to changes in land tenure. e observed gradual increase in productivity was associated with increas- ing land tenure insecurity in Finland (Ar- ticle IV), and an emerging question is how pathways out of the low productivity trap can be found (links 4 in Figure 2). One might also argue that do we need to battle against the low productivity trap? As Pend- er (1998) pointed out, the downward spiral might be a natural shift from one type of capital to another to increase income and 16 MTT SCIENCE 4 welfare at the farm level. It is true that land leasing is economically rational for a farm- er. In Finland, land rents have been mod- erate, and because the overexploitation of leased lands is not regulated, land leasing has become even more attractive. However, this does not maximize social welfare. Max- imization of social welfare calls to address questions related to pathway out of land tenure insecurity problem. ese questions are addressed by link 4 and the land sale box in figure 2. An empirical study is pre- sented in Article III. Linkages at the farm level are routed back to land tenure, because profitability at the farm level is determined by prices, subsi- dies and the productivity at which inputs are transformed to outputs. Prices and sub- sidies are to a large extent exogenous for farmers. However, farmers usually have the potential to improve productivity at the farm level if prices, subsidies and tenure- systems provide economic incentives to do so. Environmental regulations and an aid system based on positive externalities from agriculture can contribute to farm produc- tivity and profitability. Internalizing the external benefits from agriculture will lead to higher long-term land improvements (Pender 1998) and thus positively break the linkage between land tenure and land degradation. us, findings in this the- sis have relevance to the design of public policies and programmes, notably the im- portance of supplying information on the farm-level implications of land tenure and land degradation. e existing literature presents sharecrop- ping as one possibility to resolve the land tenure insecurity problem and escape the low productivity trap (Huffman and Fuku- naga 2007). Sharecropped fields have been found to produce even higher yields than sharecropping tenants’ own plots (Kassie and Holden 2007). However, sharecrop- ping is not a part of agriculture or tradi- tion in Scandinavian countries. It is also undefined who is the real holder of subsi- dy entitlements under the CAP in the EU when land is sharecropped (Swinnen et al. 2009). e landlord-tenant relationship is typically a complex informal contract con- forming with expectations, trust and rep- utation based on local traditions. In some cases, contracts are also linked to a fear of eviction. Finnish land lease legislation reduces the possibility of eviction, but at the same time it reduces the possibilities of long-term contracting. In land leasing, land tenure contracts longer than ten years are prohibited by law (Maanvuokralaki). is law dates back to the Russian revo- lution, when Finnish crofters where freed from the dictating command of large land- owners. To prevent landlordism, long-term land tenure contracts were then prohibit- ed, and they still are. Almost half of the farmland in industrial- ized countries is no longer owned by farm- ers (Ryan et al. 2001). As many landown- ers have given up farming, some of them have left their land idle, while others have leased their agricultural land to other farm- ers. ese facts make presented framework of the study feasible also in other industr- ialised country conditions. Objectives of the study1.3 e overall objective of the study report- ed in this thesis is to empirically analyse the connection between land leasing and land degradation in a well-developed agri- cultural society. ere has been consider- able interest in studying linkages between natural resource management and poverty in developing countries. However, studies concentrating on interactions between in- stitutional settings, natural resource man- agement and agricultural productivity in developed countries are, with few excep- tions, lacking. Moreover, mostly due to the lack of research based empirical evidence, there is currently no consensus on the im- pact of land tenure on land management and land degradation, or vice versa. MTT SCIENCE 4 17 Agricultural and rural systems involve high- ly dynamic and complex interactions be- tween people and soil, land management, and institutional and political components that are involved. Institutions and politi- cal components might conflict and hamper goals to improve land quality. In design- ing policies to concurrently achieve these objectives, a better understanding of their linkages is needed. Moreover, evidence on the context-specific nature of land tenure insecurity and its relationship with land degradation and productivity is required to guide policies. I hope that the results of this study can be of help for the poli- cy planners in formulating policy options to resolve the land tenure insecurity prob- lem in Finland. Outline of the thesis1.4 is thesis presents four different approach- es, reported in four separate articles, to an- alyse the economics of land tenure insecu- rity and its implications. Both normative mathematical programming and positive statistical approaches are used. Normative modelling provided a better understanding of the economics of land tenure insecuri- ty, while statistical approaches were used to test for the existence of these relations. In the first article, the land tenure insecu- rity is analysed. e objective of this arti- cle is to find out whether tenure insecurity has lead to a decline in long-term land im- provements (liming and phosphorus ferti- lization) under the Common Agricultural Policy (CAP) and Nordic production con- ditions in European Union (EU) coun- tries such as Finland. Soil pH, in partic- ular, was found to be significantly lower in land cultivated under a lease contract as compared to that owned by the farmers themselves. A large and unique dataset cov- ering the years 1998–2000 was obtained from farms in the Farm Accountancy Data Network (FADN) and was complemented by information on land parcels and land quality. e data on land parcels and their ownership status were provided by the In- formation Centre of the Ministry of Agri- culture (TIKE). Data on the indicators of soil quality, phosphorus (P) status and soil acidity (pH) were obtained from Viljavuus- palvelu Oy. e second article examined the dynamics of liming and phosphorus fertilisation in Finland in order to maximize tenant’s prof- its. e dynamic optimisation problem was solved by a stochastic dynamic program- ming routine with known parameters for one-period returns and transition equa- tions (Bellman 1957). Empirical parame- terisations of crop growth and carry-over functions represented Finnish soil quali- ty and production conditions (Kemppai- nen et al. 1993, Saarela et al. 1995, Black 1993). e decision rules were solved for alternative likelihood scenarios concern- ing the continuation of the fixed-term lease contract. e results suggested that as the probability of non-renewal of the lease contract increases, farmers rapidly re- duce investments in irreversible land im- provements and, thereafter, yields gradu- ally decline. While articles I and II suggest that land tenure insecurity has resulted in the ne- glect of land improvements, the third arti- cle analyzes whether these challenges could be resolved by a tax policy that encourag- es land sales. A mail survey was used to ac- quire data on landowners’ sale preferences, with a questionnaire being mailed to the fi- nal sample of 5 762 landowners. e data indicated that real estate tax and a tempo- ral relaxation of capital gains taxation of- fer potential for the policies aiming at the restructuring of land ownership. Howev- er, a large number of landowners (84%) in the sample were unwilling to sell their land even as a response for temporary tax shields. Potential sellers who could not be revealed by traditional logit models were identified using the latent class approach (Vermunt and Van Dijk 2001). 18 MTT SCIENCE 4 Earlier studies have shown that Finnish agriculture is suffering from low produc- tivity (Sipiläinen et al. 2008). To identify aggregate-level connections between the rapid increase in land tenure and chang- es in productivity in Finnish arable farm- ing, the productivity trend of Finnish gain farms is analysed in article IV. e nature of grain farming in Finland challenges the methodology to simultaneously take into account the general trend and annual var- iation in total factor productivity (TFP). Our analysis applied production function approach and two different models to cap- Results and discussion2 ture technical change. In the time trend model (TT), the trend variable was used as a regressor along with the input varia- bles. It was a proxy variable representing the rate of technical change or the shift in the production functions over time, and produced smooth technological changes. In the general index model of Baltagi and Griffin (1988), the trend variable t was re- placed by a vector of dummy time varia- bles. A very long time series (30 years) of data was also used to avoid biased predic- tions from short-term changes in produc- tivity trends. Land tenure insecurity in 2.1 Finland e effect of land tenure on stock vari- ables describing the quality of land was tested in article I. Stock variables where used because the data on land improve- ments were unobserved as control variables so that they could be separately identified for farmer-owned and leased land. ese stock variables were soil test P (STP) and soil pH. ey were separately observed for each plot, either farmer-owned or leased. Other plot specific characteristics where also controlled. e soil pH in Finland is higher in the South4 than in the North, but the differ- ence in mean soil pH between owned and leased land was found to be similar in both regions (Table 1). e mean soil pH was estimated to be 0.2 pH units higher for land owned by the farmer than that cul- tivated under a lease contract. In each soil group5 in the two regions, the mean soil pH for the owned parcels was higher than that of the leased parcels. However, the endpoints of the 95% confidence intervals for the mean implied that the true means of both land tenure statuses could be with- in the satisfactory class6. 4 e South covers the CAP subsidy areas A and B; the North covers the CAP subsidy area C. See CAP subsidy areas (‘tukialueet’ in Finnish) on www. mmm.fi/attachments/maatalous/tuet/5l3gtq2oH/ Paatukialueet__kartta.pdf 5 Soils are classified according to the concentration of soil organic matter (SOM) and the particle-size distribution of mineral soils. Organic soils contain >20% of SOM and mineral soils contain < 20% of SOM. e textural class and the categories of SOM (SOM1 < 3%, SOM2: 3–6%, SOM3: 6–12%, SOM4: 12–20%) of mineral soil classes are deter- mined by trained experts using finger assessment. 6 Soil characteristics are used in the interpretation of soil pH and STP results. is means that a giv- en pH or STP value, measured in the laboratory, can result in different fertility classes, depending on the soil class and the content of SOM. e results of both pH and STP are divided into seven classes as follows: poor, rather poor, fair, satisfactory, good, high and excessive. ese classes guide farmers in farming practises. MTT SCIENCE 4 19 ere was no evidence that the observed difference in pH would depend on parcel size, parcel distance from the compound of the tenant, or on any farm character- istic available in this study. Nevertheless, the soil pH was predicted to increase as parcel size increases and to decrease with parcel distance, independently of the land tenure status. As earlier pointed out, lim- ing, which increases soil pH, is one of the most important long-term investments in sustainable agricultural production in Fin- land. is is also supported by the results. In the South, which is the most produc- tive region for arable cropping, the best field plots i.e. the largest parcels, close to the farm compound have highest pH val- ues. In earlier studies, field plot size and distance from the compound have been shown to be the most critical measures of the productive value of a field plot in agri- cultural production. Above signals that leased field plots are not considered as important in long-term pro- duction plans as owned field plots, and uncertainty over land tenure has signif- icant implications for the management of leased plots. From a statistical point of view, the above-mentioned relationships between soil pH, parcel size and parcel dis- tance from the farm compound are chal- lenges to be considered when outcomes from land tenure insecurity are measured. Because of the lack of data, these challeng- es are typically ignored in studies on land tenure insecurity and effects of land tenure insecurity could not be revealed. Holden et al. (2009) presented one of exceptions to this baseline. We obtained slight evidence of a differ- ence in mean (STP) between owned and leased parcels such that soil test P values where higher on farmer-owned plots than on leased plots. e evidence was some- what stronger in the North than in the South, where in each soil group the mean STP of the owned parcels was higher than that of the leased parcels. However, in the South the estimated means for both land tenure statuses were within the satisfac- tory fertility class, i.e. in the target range. Lichtemberg (2007) used our result for STP in the South to argue against the hy- pothesis of land tenure insecurity. is is not exactly correct, because the findings from the North indicate interaction be- tween land tenure status and production line. is signals that the difference be- Table 1. Comparison of parcels cultivated by landowners and tenants with respect to covariate-adjusted mean soil pH in the South and North. a) Region / Tenure Number of parcels Adjusted mean 95% confidence interval for the mean Difference between the means Standard error of the difference P-value South b) All soils Own 210 6.2 (6.1–6.4) 0.22 0.05 < 0.0001 Leased 101 6.0 (5.9–6.2) North b) All soils Own 198 6.0 (5.9–6.2) 0.20 0.06 < 0.005 Leased 129 5.8 (5.7–6.0) a) e adjusted means are predicted values obtained from the estimated model evaluated at the two hectare parcel level, located at a distance of one kilometre from the compound. b) South covers the CAP subsidy areas A and B; North covers the CAP subsidy area C. 20 MTT SCIENCE 4 tween owned and leased parcels is smaller in livestock production than in crop pro- duction. e results suggest that P supplied in manure is quite evenly distributed on owned and leased parcels. More land may have been leased on livestock farms in the South, where pig farming is concentrated, particularly to obtain a sufficient land area for manure management. In these cases, the excessive amount of manure and phos- phorus contained in it becomes a waste product, and livestock production farms seem to equalize the STP values between leased and owned land in the South. e results indicate that a delay in land improvements, as indicated by the subop- timal stock variable values unobservable for non-agricultural landowners, does not yet show in the yield potential of land to the extent that it can be visually detected, even though the effect is statistically signifi- cant. is provides an opportunity for lease farmers to continue the overexploitation of leased lands. However, the productivity of arable farming might be hampered. Lease holder’s profit 2.2 maximization and land tenure insecurity e Finnish application of the tenants’ di- lemma concerning land improvements us- ing a recursive dynamic programming al- gorithm method in the context of grain farming is presented in Article II. is methodology is strongly based on local norms and yield responses, which have been extensively studied in field trials in Finland. e optimisation problem faced by lease farmers in liming and phosphorus appli- cation was formulated as a recursive finite- horizon dynamic programming problem that was numerically solved by iterating Bellman’s equation (Bellman 1957). Spring barley was used as an example, because it is the most widely cultivated arable crop in Finland. e optimal value function was constrained by transition equations, which were a function of the amount of lime or phosphorus used. Transition equations de- termined the connection between the cur- rent state and the control and the state in the next period. e optimal solution was estimated as a function of the initial state. is specification generalised the models presented in Kennedy (1986), where the initial state (as a proportion of the earli- er period’s application) and current appli- cation have a unique yield response. e model also exhibited the feature of stochas- tic renewal of the leasing contract. is is to my best knowledge a new feature, and implies the exogenous nature of the renew- al of land lease contracts. Land tenure insecurity constrains farm households to increase net returns through investment in land improvements by mak- ing the time horizon and, thus, the future income stream uncertain. e length of the lease contract clearly has a direct effect on the net present value of land improve- ments for the farmer in the case of a fixed probability of contract renewal. However, we adopted a more realistic approach in which the volatility of renewal probabili- ties was taken into account. Liming is to some extent lumpy as it is ex- pensive to distribute a small amount of lime. Furthermore, farmers usually do not have the necessary equipment and con- tractors are used to spread the lime on the fields. us, if the initial soil pH at the be- ginning of a five-year land lease contract is within the biological target range (5.8), which was the case in the simulations, land tenure insecurity does not make a differ- ence in the optimal liming rules within such short lease contracts. It does not pay to apply lime on land with a pH level ex- ceeding 5.4, except when the access of the farmer to the land is certain to continue, either through repeated and secure con- tract renewals or land ownership. In this case, the decision rule converges with the optimal behaviour around the long-run MTT SCIENCE 4 21 equilibrium without land tenure insecu- rity, which is also optimal from the socie- ty perspective. If land sale does not occur and land leasing continues, the initial pH level at the begin- ning of the upcoming land lease contract decreases. e decisions to lime then vary according to the uncertainty over lease re- newal. As usually noted, the duration of the contract also plays an important role in the liming decision, but this was not explicitly examined in Article II. In Arti- cle II we concentrated on the sequence of fixed-duration, five-year land leasing con- tracts and the consequences of the renew- al probability. If lease contracts are longer, the farmer can collect a larger income from the land improvements, and the threshold level determining the liming decision of landowners and tenants does not decrease to as low a level as in the case of five-year contracts. See Myyrä and Pietola (2005) for complete results from the comparison of five- and ten-year contracts. If the contract is going to expire with cer- tainty, it pays to lime at the beginning of a five-year contract if, and only if, the soil pH is below an extremely low value of 5. If the odds are slightly in favour of contract renewal (P = 0.6), it pays to lime at the be- ginning of the five-year contract if the soil pH is below 5.2. e risk-free long-term equilibrium is to maintain soil pH above 5.4, representing a “fair” soil pH status in the soil type studied. When contract re- newal is likely, e.g. P > 0.5, it is still advan- tageous to maintain the soil pH above 5.2. Optimal timing is to apply lime immedi- ately after the new contract is signed. Nev- ertheless, when the likelihood for contract renewal decreases and the odds are in fa- vour of contract termination (P < 0.5), the soil pH is allowed to decrease below 5.2, which most commonly represents the fer- tility class “rather poor” or “poor”7. e study of phosphorus application un- der land tenure insecurity in Article II fo- cused on the difference between two steady state equilibriums. e first represented the case where no land tenure insecurity exists, i.e. the land is owned by the farmer or the lease contract is repeatedly renewed with certainty. e steady state equilibrium for phosphorus application in barley produc- tion was estimated at 18 kg/ha, which is slightly exceeds the environmental recom- mendations, implying that environmen- tal regulations limit phosphorus applica- tion by owner farmers. e likelihood for the non-renewal of the lease contract has a considerable effect on the optimal phos- phorus fertilisation, particularly when the contract is approaching the renewal date. If, for example, the odds are slightly in fa- vour of contract renewal (P = 0.6), the op- timal phosphorus application was found to remain between 10 and 11 kg/ha un- til the third year before the renewal date. is level of application is as much as 39 to 44% lower than the long-term equilibri- um without land tenure insecurity. ere- after, and towards the end of the current contract, the phosphorus application drops to 8–9 kg/ha. If the contract is to expire with certainty, phosphorus application rap- idly declines towards 2 kg/ha as the expi- ry date approaches. Clearly, the environ- mental regulations do not hinder annual phosphorus fertilization decisions by lease farmers, particularly when the contract is approaching its renewal date. e key interest in Article II lay in the long-term effects of the current land ten- ure system, which is characteristic of the Finnish agricultural sector. e case of pH was discussed above, and the details of the phosphorus stock measured as soil test P (mg/l) on farm land are described in Fig- ure 3. Due to continuous lease farming, a gradual decrease in soil phosphorus stock has occurred. When there is no land tenure insecurity, the steady state for soil phos- phorus, i.e. the long-term equilibrium that maximises the value function, is 8.5 mg/l, 7 e class limits depend on the particle-size distri- bution and the organic matter content of the soil. 22 MTT SCIENCE 4 which is substantially below the amount needed to produce the maximum yield (13 mg/l). If uncertainty over the continuation of each five-year contract is large (the prob- ability of contract renewal, P = 0.2), the steady state soil phosphorus is estimated at 5.5 mg/l, which definitely represents a low- er soil phosphorus status than that required for the maximum yield. is extremely low level of soil phosphorus describes the over- exploitation of land by tenants. e economic loss due to a decreased pH and STP stock arises from several reasons. First, liming is a necessary land improve- ment in Finnish conditions to maintain the soil pH. A decrease in pH impedes the up- take of almost all nutrients needed by grain (Kemppainen et al. 1993). Second, changes in the phosphorus stock have implications for the fertilising behaviour of tenants. Un- der a decreased phosphorus stock, annual phosphorus fertilization by lease holders becomes a difficult decision and even more dependent on the probability of contract renewal. Because of the low phosphorus stock there is an incentive to use high lev- els of phosphorus fertiliser at the beginning of the lease contracts. ird, as our results from the liming and pH model show, lime is applied at the beginning of every second or third lease contract and the overall pro- ductivity of the land declines. e yield po- tential is no longer large enough to trans- form these phosphorus fertilization peaks at the beginning of the lease contract to improved yields. Furthermore, some of the yield potential might become underutilized because of the low levels of phosphorus fer- tilization at the end of lease contracts. is type of phosphorus fertilization peaking at five-year intervals is not efficient from an environmental perspective. ree interesting phenomena could be identified from the above results. First, current cash lease contracts give a clear in- centive for a tenant to transform land cap- ital, i.e. land fertility stock to other forms of capital. Second, cultivation practises are significantly driven by the economic and institutional settings. is might lead to major insufficiencies in plant nutrition and thus hinder the productivity of leased land. Figure 3. Converged values of easily soluble P, conditional on the probability (Prob) that the five-year leasing contract is renewed. If prob = 1 (Prob = 0), the contract is renewed (terminated) with certainty. ?????%?????%?%??1? ? ? 9 8 7 6 5 4 3 2 ?????3 ?????20? ?????208 ?????206 ?????204 ?????202 ??1? MTT SCIENCE 4 23 Phosphorus fertilization is just one exam- ple of these trends. ird, the productivi- ty of farmland is reduced and maintenance of the food supply through high yields is jeopardized. is also hampers the goals of environmental programmes, because the yield is the only positive way to collect nutrients from agricultural soils and pre- vent runoff. Taxation and land sale2.3 In Article III possible effects of taxes as pol- icy instruments to promote farmland land sales is studied. From the land sales sta- tistics and studies we find that only 1.8% of the total field area is traded annually (Pyykkönen 2006). is raises an impor- tant question concerning how tax policies, for example, could be used to encourage land sales to solve the problem of the low productivity trap. Some earlier studies have also confirmed that the land use intentions of landowners can be affected by policy in- struments. However, earlier studies have also demonstrated that landowners may be reluctant to sell their agricultural land. Numerous reasons have been presented, but in many respects the type of landown- er and the property characteristics associ- ated with unwillingness to sell remained unknown. In Finland, some institutionally available policy instruments could be used to en- hance land markets. Temporary relaxation of taxation on the proceeds of a farmland sale (TTF) would be one equitable op- tion, because it is open to all landowners and quite light taxes are currently collect- ed from capital gains in land sales. Anoth- er alternative is to introduce a real estate tax (RET) on farmland. RET affects all landowners, but it could be conjectured to trigger8 land sales amongst landowners whose land use does not generate land rent that exceeds the tax. A prerequisite for suc- cessful policies is a thorough knowledge of landowners’ preferences and willingness to take part in markets because of these types of tax programmes. However, such knowl- edge is currently lacking. To study landowners’ preferences over land sales, a large survey was conducted on a sample of landowners, including both ac- tive farmers and passive landowners, in 2006. e sample was selected from the register of the Finnish tax administration based on the data for 2004. A questionnaire was mailed to the final sample of 5 762 landowners and a total of 2 684 observa- tions were obtained. e questionnaire focused on a single holding of farmland owned by the responding landowner. All land sale models, including past sales, sales under TTF and RET as well as classifica- tion of land owners to sellers and less prob- able sellers, were fitted to this dataset. Based on past sales, the standard logit mod- el identified only some landowner char- acteristics related to sale. Unsurprisingly, unemployment explained past sales. e acute need for income from land sale for consumption was a logical and plausible explanation. A somewhat unexpected re- sult was the finding of regional differences in past sales. e probabilities for past sales were higher in the western part of Finland and lower in the northern part. However, these patterns could be related to cultur- al differences, and to general expectations concerning agriculture and land markets, which are decreasing in the north and in- creasing in the west. Although we lost a large number of obser- vations because landowners found it dif- ficult to report their future intentions to sell their land, the logit models performed somewhat better when future land sale in- tentions were investigated. We observed that earlier sales were the dominant pre- dictor of future sales and that 5.8% of land 8 Holt, C. and Shelton, J. 1962. e lock-in effect of the capital gains tax. National Tax Journal 15: 337–352. 24 MTT SCIENCE 4 owners had an experience of earlier land sales. is result was consistent with pre- vious studies (Ajzen 2002) and confirmed that the same factors explain both past and future sales. However, the logit models in- dicated that these factors are rare and dif- ficult to identify. We found some evidence of a connection between female gender, a higher age and lower education, as well as landowner objectives9 in selling land. Despite the association between certain so- cioeconomic variables and the probability of future land sales, the power of the log- it models to correctly predict actual sellers was weak. Using a cut-off point of 0.50, as in standard in computing predictions in these types of probability models, the number of correctly-classified sellers was under 20% of all observed sellers within the sample. By lowering the cut-off point the sensitivity could be improved, but a loss of specificity was inevitable. is re- sult inspired us to try some new method- ologies not yet applied in this field of eco- nomic analysis. To gain a further insight into the selling intentions we applied a latent class model. An interesting feature of latent class model- ling is the ability to use statistical criteria to measure the number of underlying latent groups. An example of latent groups could be those value-based objective groups that we used earlier. e series of fitted latent class models and the applied Bayesian in- formation criterion (BIC) confirmed the presence of two classes of farmland owners concerning land-sale preferences. Class 1 consisted of owners who were less probable sellers in the coming five-year period. is class comprised 84% of all landowners. Class 2, comprising 16% of landowners, had future intentions to sell land, particu- larly if a tax policy would be implemented. At this point it has to be mentioned that the number of classes in our model is not necessary limited to two. Behind the ob- served variables, a number of unobserved variables may exist and parameters of the regression model explaining land sale de- cisions differ across unobserved classes. If these underlying landowner classes are not recognised, serious problems due to omit- ted variables are faced. e effect of the studied policy measures (TTF and RET) differed significantly be- tween the observed landowner groups. In the model containing the dichotomous RET variable, higher tax levels were as- sociated with higher sale preferences, but the effect was stronger in the class of land- owners with intentions to sell land (class 2). e difference between the landown- er classes was even stronger on the case of TTF. In class 1, sale intentions were not significantly affected by the TTF policy, but in class 2 TTF had a positive and sig- nificant effect on sales. e introduction of substantially high RET levels (from €10 per ha to €70 per ha) was found to increase the selling probability in Class 2 from 17% to 70%. Almost the same increase in selling intention could be achieved by TTF, which increased the sell- ing intention by 53% in class 2. Based on these results, TTF was concluded to be a more attractive policy if an increase in land sales was seen as one possible solution to the current land tenure insecurity prob- lem in Finland. Some of the observable land owner’s char- acteristics where connected to fitted latent classes. However, the earlier sale behaviour remained as the best predictor for the fu- ture sales. Characteristics like land owner- ship objective group, the region of the farm and the perception of the field prise gives tools for future research to predict farm- land markets. But, it is not possible to dif- 9 Five value-based objective groups (agricultural earners, indifferent owners, immaterialists, part-time farmers and agricultural earners) were constructed in a paper by Pouta, Myyrä and Hänninen. e paper has been submitted to the journal Society and Nat- ural Resources. MTT SCIENCE 4 25 ferentiate the taxation between land own- ers based on these results. Long-term productivity 2.4 changes in grain production Changes in agricultural productivity have been of continuous interest to agricul- tural economists in Finland (Ihamuoti- la 1972, Hemilä 1982, Ylätalo 1987, Ry- hänen 1994, Sipiläinen 2008). e first studies on TFP changes in Finland by Iha- muotila (1972) indicated a rate of growth of TFP between 2.5–2.7% per year from 1950–1969. Ylätalo (1987) estimated the annual TFP change in Finnish agriculture from 1961–1984. e results indicated TFP growth between 2.9% and 1.5%, de- pending on the income shares of labour and capital. e rate of increase in TFP im- proved towards to the end of the research period, being 3.8% from 1975–1984. Si- piläinen (2008) showed that productivi- ty growth on Finnish cattle farms faced a slowdown in the mid-1990s at the time of EU accession, but have accelerated since then. During the 1990s the average in- crease in productivity on cattle farms was estimated at 1.0% per year. Long term pro- ductivity development is studied in last study of this thesis, Article IV. In Article IV, the average production func- tion is estimated. By assuming allocative efficiency the estimation of production function can be concentrated to derive technical change and the scale effect on productivity growth. By this technical de- composition, no detailed price informa- tion is needed, although it is assumed that farmers face equal prices in order to be able to estimate changes in TFP. e main chal- lenge is thus to estimate a suitable produc- tion function. e simultaneity problem caused by the used endogenous explana- tory variables in the production function could not be avoided. e one way to go around the simultaneity problem could have been the usage of dual approach, but for that there where not enough farm spe- cific price information. With two separate models we showed that the annual variation in TFP growth rate in Finnish grain farms was large and captured by the general index model (GI), because this model allows more flexibility with re- spect to year-to-year changes in produc- tivity. However, it might be impossible to capture the general trend in TFP based only on general index model results. To test larger trends we used the time trend mod- el (TT). is issue was topical during the first years of EU membership, when land leasing rapidly increased. As Figure 4 illus- trates, the 1994 harvest year was reasona- bly good. However, during the following years from 1995 to 1999, the first five-year period under the CAP, weather conditions were poor and TFP collapsed. At time of accession to the EU, there was considerable debate over quasi or appar- ent farming. is was because of the sharp decrease in the marginal product value (MPV) and increasing land leasing, both of which reduced the incentives for input us- age. Despite clear changes in the operating environment in farming, it still is difficult to find changes in the productivity trend, even if we have such a long time frame and advanced econometric methods. Our results from article IV indicated that total factor productivity (TFP) increased in 1976–2006 by 0.6% per year in the GI model and 1.7% per year in the TT model. Results from the TT model also evidenced that the increase in productivity was hin- dered by increasing land tenure and the policy changes introduced in 1995. e cumulative increase in TFP indicated by TT model over the study period was at the same level as the largest yearly changes in TFP estimated by the GI model. e re- sults are presented in Figure 4. 26 MTT SCIENCE 4 Based on our experience, the time trend model is not flexible enough to capture the annual variation in Finnish grain farming, and in this sense results gained from the general index model are preferable. Balta- gi and Griffin (1988) also preferred the GI model, since it closely follows the chang- es in Divisia indices. Kumbhakar et al. (1999), however, presented some contra- dictory views on the TT model. Farming is a long-term operation and land improvements have a long payback time. Consequently, it is natural that sharp economic changes generate only gradual changes in land productivity. However, it could be argued that the increase in TFT of grain farming could have been faster with- out combined effects from land tenure and changes in policy regime. According to Fuglier (2008), agricultur- al productivity increased in Western Eu- rope by 1.6% per year on average from 1970–2006, while it increased in Eastern Europe by only 0.6% per year. is is in- teresting, because property rights and land reforms are known to be among the ma- jor questions in Eastern European agricul- ture. A slow privatization process prolongs the uncertainty of property rights. As long as property rights are uncertain, markets cannot develop and the disinvestments in agriculture continue, including a reduc- tion in investment in land (Macours and Swinnen 2000). Figure 4. Annual change in total factor productivity. TT = Time trend. GI = General index. 19 76 19 78 19 8 0 19 8 2 19 8 4 19 8 6 19 8 8 19 9 0 19 9 2 19 9 4 19 9 6 19 9 8 2 0 0 0 2 0 0 2 2 0 0 4 2 0 0 6 TT GI 70 50 30 10 %10 %30 %50 MTT SCIENCE 4 27 Conclusions3 economic optimisation model, construct- ed in a dynamic programming framework. us, the lacking causality was modelled and calibrated to Finnish conditions. e tenants’ dilemma over land improve- ments is important and a concern of the increasing field area in Finland. e most important new feature in the dynamic pro- gramming application was the recognition of the stochastic nature of the duration of lease contracts and explicit modelling of the probability of renewal of land lease contracts. e results suggested that un- certainty over the renewal of the lease con- tract sharply reduces incentives for farmers to invest in maintaining pH and STP. Be- cause soil pH and STP are necessary factors to obtain good yields, the yields and land productivity gradually decrease when these soil improvements are not carried out. e results signal that in Finnish land ten- ure market conditions, where land tenure contract terms are limited, the reputation of farmers gained through land improve- ments does not resolve the problem of land tenure insecurity. It does not pay to make land improvements to build up a long-term land tenure relationship with the landown- er. e reasons for this remained unclear, but it could mostly be because landowners are no longer attached to agriculture and the productive value of agricultural soils. If there is no possibility or legal incentive to establish conditions for long-term planning by giving farmers and landowners possibil- ities to make arrangements longer than ten years ahead, and thereby to reduce uncer- tainty, it might be beneficial for society to relax land market constraints, and the pos- sible lock-in effect, through other policies such as taxes. Property rights are fundamental to economic performance and efficien-cy analysis. Land improvements are necessary, but an insufficient condition in arable farming to improve performance. ese fundamental phenomena are com- bined in this thesis. Certain conditions of land acidity, measured by pH, and availa- bility of phosphorus, measured by STP, are essential to obtain maximum crop yields. is study described how these conditions are most unlikely to be met under the cur- rent land tenure system in Finland, where farmers are confronted with substantial land tenure insecurity. e results under- line that the low productivity trap caused by land tenure insecurity, land degrada- tion and declining productivity can also occur in highly modernized, developed and rich countries. Under these conditions, the main driving factor for the low productivi- ty trap is uncertainty. Uncertainty over the continuation of short-term cash lease land tenure contracts restricts farmers in their ability to discount future income streams from current land improvements. Both the soil pH and STP were shown to be lower on leased land than on the land owned by the farmer in Finland. e limitation of the data set to cross-sec- tional observations of plot characteristics and stock measures of soil pH and STP limits the ability to assert and explicitly test for the causal relationship between land tenure and land improvements, i.e. corre- sponding control variables which in this case are liming and P fertilization. Nev- ertheless, I refer to the observed statistical relationships as “causality”. Obviously, a causal relation could be only justified based on time series data. However, such a data set was unavailable for this study. e caus- al relationship was analysed in a separate 28 MTT SCIENCE 4 One pathway in agricultural policy might be to encourage landowners to take full ad- vantage of the market value of agricultur- al land. Relaxation of the capital gains tax on land sales showed some potential along these lines. However, 86% of landowners did not consider the sale of their land as a plausible option. It is evident that the productivity of leased land in Finland is declining. In future years the land productivity will decrease to a point where we will have to consid- er if it is feasible or beneficial for socie- ty to try to maintain the entire field area under agricultural production. is is es- pecially so in those areas where land leas- es have reached the level of 0 €/ha or mar- kets are expecting land to be converted to some other usage than agriculture. Because landowners are not willing to sell land to active farmers, other interests of the land- owner as well as society at large have to be accounted for. A change in land usage of fields with the lowest productivity is one of the fastest ways to improve the general pro- ductivity of Finnish agriculture. is study revealed that land tenure and low produc- tivity could be connected, and potential ar- eas for new agricultural policy indicated by the history of land tenure. e possibility to reallocate subsidy entitlements for the most productive farm lands might be one direct way to improve land allocation and productivity. e decreasing trend in productivity on leased land is natural and occurring un- der current and well-established land ten- ure contracts. Connected to the market environment, these contracts provide in- centives for farmers to transform land im- provements, i.e. inbuilt productivity stock, to cash. e markets do not penalise farm- ers if they do not invest in land and this way try to ensure the continuum of the land tenure contract. Although natural, this trend is alarming from the perspec- tive of productivity development in arable crop farming. Agriculture is a long-term operation, and the current, partially “home made” land tenure institutions in Finland are hindering the survival of Finnish agri- culture under international competition. Land institutions and land tenure ques- tions are essential in agricultural and rural policies on all levels, from local to inter- national. Land ownership and land titles are commonly tied to fundamental politi- cal, economic and social questions. A fair resolution calls for innovative and new so- lutions both on national and internation- al levels. However, this seems to be a prob- lem when considering the application of EU regulations to member states inher- iting divergent landownership structures and farming cultures. e contribution of this study is in describing the consequenc- es of fitting EU agricultural policy to Finn- ish agricultural land tenure conditions and heritage. MTT SCIENCE 4 29 Yhteenveto (summary in Finnish)4 toviljelyn tuottavuuskehitystä tarkasteltiin pitkällä, yli 30 vuoden mittaisella paneeli- aineistolla. Käytetyt ekonometriset mene- telmät ja tuotanto taloudellinen lähestymis- tapa kytkevät tutkimuksen maatalouden tuotantotaloudelliseen viitekehykseen. Pellon hallintaoikeuden epävarmuuden li- säksi pellon kasvukunnon rapautumista on selitetty kansainvälisessä kirjallisuudes- sa monilla tekijöillä. Köyhyys, asutuksen paine, huonosti toimivat tai olemattomat markkinat panoksille, tuotteille ja maata- louden ulkoisvaikutuksille ovat tekijöitä, joita pidetään syynä pellon sadontuottoky- vyn alenemiseen. Näiden tekijöiden mer- kitys vaihtelee voimakkaasti eri alueiden välillä, mutta mikä tahansa niistä voi lau- kaista matalan tuottavuuden loukun. Edel- lä mainittujen tekijöiden vaikutuksen eroja voidaan selittää esimerkiksi maanomistuk- sen rakenteella, alkuperäisillä tai ikiaikaisil- la maaoikeuksilla sekä sosiaali silla raken- teilla ja normeilla. Suomalaiset olosuhteet ovat erikoiset verrattaessa nii tä aikaisem- paan kansainväliseen kirjallisuuteen mata- lan tuottavuuden loukusta, sillä toimivien markkinoiden ja kehittyneiden sopimus- käytäntöjen on arvioitu ratkaisevan nämä ongelmat. Tulokset kuitenkin paljastavat, että vuokrapeltojen sadontuottokyky on Suomessa jäämässä jälkeen muiden mai- den peltojen vastaavasta. Aiempi tieteellinen keskustelu matalan tuottavuuden loukusta painottuu kehi- tysmaihin. Ehkä juuri tästä johtuen riit- tävän informatiivisen kvantitatiivisen em- piirisen aineiston käyttö tutkimusaiheen tutkimuksessa ei ole lopultakaan ollut ko- vin yleistä. Välillisiä perusparannuspäätök- siä, kuten eroosiota estäviä toimia, on tut- kittu paljon, mutta todellisia omistajan ja vuokraviljelijän viljelemien peltojen sadon- tuottokyvyn eroja ei ole useinkaan tilas- Maatalous on pitkäjännitteistä toi-mintaa. Pellon perusparannuksia suunnittelevan maanomistajan suunnittelujänne on pitkä, jopa yli suku- polvien ulottuva. Sen sijaan vuokraviljeli- jän suunnittelujänne on perusparannusten osalta lyhyt. Usein väitetäänkin, että vuok- raviljelijöillä on taipumus perusparannuk- sia rapauttavaan viljelyyn. Perusparannus- ten huomioonottaminen antaa haasteita vuokrasopimuskäytäntöjen kehittämiseen. Vuokrasopimusten ja tukijärjestelmän sekä maatalousmarkkinoiden yleensäkin tulisi olla riittävän tehokkaita antaakseen vuok- raviljelijälle mahdollisuuden ja kannus- teet myös vuokrapeltojen pitkäjänteisen sadontuottokyvyn ylläpitämiseen. Pelto- jen hyvä sadontuottokyky ja maan tuotta- vuus ovat koko yhteiskunnalle monestakin syystä tavoiteltavia asioita. Vuokrapeltojen hallintaoikeuden epävarmuuteen keskitty- vä tutkimus on juuri nyt erittäin ajankoh- tainen, sillä pellon vuokraus on yleistynyt 1990-luvulla ja sen jälkeen nopeasti. Toi- saalta maatalouden tuottavuuskehityksen, ja erityisesti kasvinviljelyn tuottavuuskehi- tyksen, on havaittu olevan hidasta. Epäil- läänkin, että osa vuokrattuina viljellyistä pelloista on ajautumassa mata lan tuotta- vuuden loukkuun, jossa sekä maanomista- jan että viljelijän kannusteet pellon kasvu- kunnon parantamiseen ovat alenevat. Tässä tutkimuksessa tarkasteltiin vuokra- viljelystä johtuvan pellon hallintaoikeuden epävarmuuden vaikutusta viljelijän teke- miin pellon perusparannuksiin. Kahdessa ensimmäisessä artikkelissa tutkit tiin pel- tojen vuokrauksen vaikutusta niiden kal- kitusten ja fosforilannoitusten mää riin ja kannattavuuteen. Kolmannessa artikkelis- sa keskityttiin havaitun pellon hallintaoike- uden epävarmuudesta johtuvan ongelman ratkaisuvaihtoehtoihin verotusjärjestelmää kehittämällä. Neljännessä artikkelissa pel- 30 MTT SCIENCE 4 tollisesti testattu. Poikkeuksen tästä pää- linjasta tekee Yhdysvalloissa ja Kanadassa havaittu eroosio-ongelma. Suomessa käy- tettävissä oleva aineistokaan ei mahdollis- ta satotasoerojen selvittämistä vuokra- ja omistajapeltojen välillä, sillä meillä ei ole puhtaita vuokraviljelijöitä. Käytettävis- sä oleva aineisto kuitenkin mahdollistaa sen, että peltojen sadontuottokykyä pel- lon viljavuudella mitattuna voidaan vertail- la. Tämä vertailu tehtiin artikkelissa I, jossa tärkeimpänä tavoitteena oli selvittää, onko vuokrapeltojen sadontuottokyky jo jäänyt jälkeen omistajapeltojen vastaavasta. Artikkelin I tulokset osoittavat, että viljeli- jän omien ja vuokrapeltojen pH:n taso ero on nyt ½ viljavuusluokkaa ja helppoliu- koisen fosforin tasoero ⅓ viljavuusluok- kaa viljelijän omien peltojen hyväksi. Pelto- jen kasvukunnon ollessa kohtalaisen hyvä ei viljavuuden tasoero aiheuta kuitenkaan vielä merkittävää eroa omien ja vuokrat- tujen peltojen satotasojen välille. Keskei- sin tulos piileekin juuri tässä. Viljavuu- den tasoeron kehitys on niin hidasta, että markkinat eivät pysty reagoimaan siihen ja vuokramiehet voivat jatkaa vuokrapeltojen perusparannuksia rapauttavaa viljelytapaa. Tästä syntyy matalan tuottavuuden louk- ku, johon reilusti yli ⅓ Suomen pelloista on luisumassa. Artikkelissa II keskityttiin siihen, ettei ka- usaalisuhdetta pellon vuokrauksen ja ha- vaitun viljavuuseron välillä ei voitu tilas- tollisesti vahvistaa Artikkelissa I käytetyn poikkileikkausaineiston perusteella. Ka- usaalisuhteen talousteoreettinen perusta ra- kennettiinkin Artikkelissa II dynaamisen ohjelmoinnin menetelmin. Ra kennetut dy- naamisen ohjelmoinnin mallit ovat maata- louden liiketaloustieteen ja erityisesti tuo- tantotalouden keskeisimpiä menetelmiä. Viljelijän toiminnan ratio naalisuuteen, kuten voiton maksimoin tiin, nojaavat mallit rakennettiin aikaisemman tutki- mustiedon varaan viljelysmaan pH:n ja fosforilannoituksen vaikutuksesta satoon. Mallien uutuutena ja samalla poikkeuk- sena oli vuokrasopimuksen jatkumiseen liittyvän epävarmuuden huomi oiminen ulkoa annettuna eli eksogeenisena. Tämä lähestymistapa lisää kansainväliseen kir- jallisuuteen aivan uuden näkökulman, sillä aikaisemmin vuokraviljelijän perus- parannuskannusteiden on nähty osittain rakentuvan viljelymaineen rakentamisen varaan. Vuokrasopimuksen jatkumi sen to- dennäköisyyttä on siis pidetty endogee- nisena. Vuokrapeltojen ja omien peltojen viljavuus eron perusteella väitteeni kuiten- kin on, että vuokrasopimuksen jatkumisen todennäköisyyden tarkastelu eksogeenisena muuttujana kuvaa paremmin pellonvuok- ramarkkinoita Suomessa. Mallin tulokset osoittavat, että vuokraso- pimuksen uusiutumisen todennäköi syy- den laskiessa vuokraviljelijät vähentävät nopeasti sekä kalkitusta että fosforilannoi- tusta. Tästä seuraa satotason asteittainen aleneminen. Artikkelien I ja II osoittaessa vuokrapelto- jen sadontuottokyvyn olevan jäämässä jäl- keen viljelijöiden omien peltojen vastaa- vasta artikkelissa III tarkastellaan keinoja ongelman ratkaisuksi. Artikkelissa tarkas- tellaan erilaisten verokannusteiden vaiku- tusta pellon myyntihalukkuuteen. Kaik- kiaan 5 762 pellonomistajalle lähetetyllä kyselyllä kartoitettiin pellon myyntiha- lukkuuteen vaikuttavia tekijöitä. Aineis- ton perusteella näyttää siltä, että pellon kiinteistövero sekä väliaikainen pellon myyntivoittoveron huojennus tarjoavat mahdollisuuksia peltokauppojen volyy- min lisäämiseen. Aineisto osoitti kuiten- kin myös sen, että 86 % pellonomistajista ei myy peltoa ehdotetuista veromuutoksis- ta huolimatta. Potentiaalisia pellon myyjiä ei voitu paljastaa perinteisellä logit-mallin- nuksella, mutta myyntihaluiset saatiin esiin Latent Class (LC)-menetelmällä. LC-me- netelmä osoittikin selkeää potentiaalia tut- kimusmenetelmänä tarkasteltaessa omis- tajien käyttäytymistä, jota ei voida kytkeä helposti mihinkään omistajia kuvaavaan sosio demografiseen muuttujaan. MTT SCIENCE 4 31 Artikkelissa IV viljatilojen tuottavuus- ke hi tys vuosina 1976–2006 estimoi- tiin kannattavuuskirjanpitotila-aineistosta maa talouden tuottavuuskehityksen ja pel- lonvuokrauksen välisen yhteyden selvittä- miseksi. Suomalaisen kasvinviljelyn erityis- piirteet ja erityisesti tuotannon voimakas sääriippuvuus haastavat tuottavuuskehi- tyksen mittausmenetelmät. Menetelmän pitää olla riittävän joustava huomioidak- seen vuotuisen vaihtelun, mutta toisaalta taustalla olevat hitaasti muuttuvat ilmiöt eivät saisi jäädä vuotuisen vaihtelun alle. Käytetyssä analyysissä nämä seikat otet- tiin huomioon käyttämällä kahta eri ta- valla teknologista kehitystä kuvaavaa mal- lia. Mallit perustuvat tuotantofunktion määrittämiseen ja siitä johdettuihin tulok- siin. Tulokset osoittavat, että vilja tilojen tuottavuuskehitys on hidastunut vuonna 1995 tapahtuneiden maatalouspolitiikan muutosten myötä. Tämä tulos oli tilastol- lisesti merkitsevä, mutta todelliselta mer- kitykseltään melko pieni. Tulos kuitenkin vahvistaa käsitystä siitä, että EU-jäsenyy- den myötä voimakkaasti lisääntynyt pel- lonvuokraus on osaltaan hidastanut maa- talouden tuottavuuskehitystä. Hyvin järjestetyt tuotantovälineiden omis- tussuhteet luovat kannusteet pellon perus- parannuksille. Pellon perusparannukset ja hyvä kasvukunto ovat välttämättömiä, mutta eivät riittäviä ehtoja kasvinviljelyn tuotta vuuden parantamiseksi. Tässä opin- näytetyössä saadut tulokset osoittavat, että Suomessa käytössä olevat vuokrasopimus- käytännöt, vuokrasopimusmal lit sekä pel- lonvuokraukseen liittyvä lainsäädäntö eivät pysty ratkaisemaan vuokrapellon hallin- taoikeuden epävarmuudesta aiheutuvia ongelmia. 32 MTT SCIENCE 4 References in southern Ethiopia. Land Economics 78: 573–590. Huffman, W.E. & Fukunaga, K. 2007. Sus- tainable land use: landlord–tenant con- tracting in the United States of America. Njas-Wageningen Journal of Life Sci- ences 55: 379–96. Ihamuotila, R. 1972. Productivity and the aggregate production functions in the Finnish agricultural sector, 1950-1969. Publications of the Agricultural Econom- ics Research Institute, Finland 25. Kassie, M. & Holden, S. 2007. Sharecrop- ping efficiency in Ethiopia: threats of eviction and kinship. Agricultural eco- nomics 37:179–188. Kataja, J-P. 2008. Sale price of the forest estate and the factors affecting forest land sale price. University of Helsinki, Faculty of Agriculture and Forestry, De- partment of Forest Economics. Mas- ter’s thesis. 61 p. Kemppainen, E., Jaakkola, A. & Elonen, P. 1993. Peltomaiden kalkitustarve ja kalkituksen vaikutus viljan ja nurmen satoon. In Finnish, with an English ab- stract: Effect of liming on the yield of cereals and grass. Maatalouden tut- kimuskeskuksen tiedote 15/93. Kennedy, J. 1986. Dynamic programming. Applications to agriculture and natural resources. Australia. Kumbhakar, S.C., Heshmati A. & Hjalmars- son, L. 1999. Parametric Approaches to Productivity Measurement: A Com- parison among Alternative Models. Scandinavian Journal of Economics 101: 405–424. Li, G., Rozell, S. & Brandt, L. 1998. Tenure, land rights, and farmer investment in- centives in China. Agricultural Econom- ics 19: 63–71. Lichtenberg, E. 2007. Tenants, landlords, and soil conservation. American Jour- nal of Agricultural Economics 89: 294– 307. Macours, K. & Swinnen, J. 2000. Causes of Output Decline in Economic Transition: The Case of Central and Eastern Euro- Allen, D. & Lueck, D. 2002. The Nature of the Farm: Contracts, Risk, and Organi- zation in Agriculture. Massachusetts Institute of Technology. Baltagi, B.H. & Griffin, J.M. 1988. A general index of technical change. Journal of Political Economy 96: 20–41. Bellman, R. 1957. Dynamic Programming. New Jersey. Black, C.A. 1993. Soil fertility evaluation and control. CRC Press Inc., Boca Raton, Florida, USA. Deininger, K. 2003. Land Policies for Growth and Poverty Reduction. A World Bank Policy Research Report. Deininger, K. & Chamorro, J. 2004. Invest- ment and equity effect of land regula- tion: the case of Nicaragua. Agricultural Economics 30: 101–116. Fraser, E. 2004. Land tenure and agricul- tural management: Soil conservation on rented and owned fields in southwest British Columbia. Agriculture and Hu- man Values 21: 73–79. Fuglie, K.O. 2008. Is a slowdown in agricul- tural productivity growth contributing to the rise in commodity prices? Agricul- tural Economics 39:431–441. Gutierrez, L., Westerlund, J. & Erickson, K. 2007. Farmland prices, structural brakes and panel data. European Review of Agricultural Economics 34: 16–179. Hardin, G. 1968. The Tragedy of the Com- mons. Science 162:1243–1248. Hemilä, K. 1982. Measuring technological change in agriculture: An application based on the CES production function. Journal of the Scientific Agricultural So- ciety of Finland 54: 165–223 Holden, S., Deininger, K. & Ghebru, H. 2009. Impacts of low-cost land certifi- cation on investment and productivity. American Journal of Agricultural Eco- nomics 91: 359–373. Holden, S. & Hailu, Y. 2002. Land redistri- bution, tenure insecurity and intensity of production: A study of farm households MTT SCIENCE 4 33 pean Agriculture. Journal of compara- tive economics 28:172–206. Mas-Colell, A., Whinston, M.D. & Green, J.R. 1995. Microeconomic Theory. Ox- ford University Press. McConnell, K. 1983. An Economic Model of Soil Conservation. American Journal of Agricultural Economics 65: 83–89. Myyrä, S., Pouta, E. & Hänninen, H. 2008. Suomalainen pellonomistaja. Maa- ja elintarviketalous 115. 182 p. Myyrä, S. & Pietola, K. 2005. Land Improve- ments under land Tenure Insecurity. The case of liming in Finland. In: 94th EAAE Seminar. 9th - 10th April 2005, Wye Campus. Imperial College London. 11 p. http://www.eaae-wye.org.uk/pro- gramme.html Myyrä, S. & Scrimgeour, F.G. 2008. Farm- land prices in New Zealand and Finland. In: Oxley, L. & Kulasiri, D. (ed.). Land, Water and Environmental Management: Integrated Systems for Sustainability, Proceedings. p. 1061–1067. Nkonya, E., Pender, J., Kaizzi, K., Kato, E., Magurura, S., Ssali, H. & Muwonge, J. 2008. Linkages between Land Manage- ment, Land Degradation, and Poverty in Sub-Sahara Africa. International Food Policy Research Institute. Research Re- port 159. 122 p. Pender, J. 1998. Population growth, agricul- tural intensification, induced innovations and natural resource sustainability: An application of neoclassical growth theo- ry. Agricultural Economics 19: 99–112. Prokopy, L., Floress, K., Klotthor-Weinkauf, & Baumgard-Getz, A. 2008. Determi- nants of agricultural best management practice adoption: Evidence from litera- ture. Journal of Soil and Water Conser- vation 63: 300–311. Pyykkönen, P. 2006. Factors Affecting Farmland Prices in Finland. Helsinki: Pellervo Economic Research Institute Publication 19. Reardon, T. & Vosti, S. 1995. Links Between Rural Poverty and the Environment in Developing Countries: Asset Categories and Investment Poverty. World Devel- opment 23: 1495–1506. Ryan, J.C., Barnard, R., Collender, R. & Erickson, K. 2001. “Government pay- ments to Farmers Contribute to Ris- ing Land Values”. Agricultural Outlook (June-July 2001): 22–26. Ryhänen, M. 1994. Input substitution and technological development on Finnish dairy farms for 1965-1991. Agricultural Science in Finland 3: 525–599. Saarela, I., Järvi, A., Hakkola, H. & Rinne, K. 1995. ”Fosforilannoituksen porrasko- keet 1977-1994. Vuosittain annetun fos- forimäärän vaikutus maan viljavuuteen ja peltokasvien satoon monivuotisissa kenttäkokeissa.” Abstract: Phosphorus fertilizer rate trials, 1977-1994. Effects of the rate of annual phosphorus appli- cation on soil fertility and yields of field crops in long-term field experiments. Maatalouden tutkimuskeskuksen tie- dote 16/95. Sipiläinen, T. 2008. Components of produc- tivity growth in Finnish agriculture. Agri- food Research Reports 116. 154 p. Sipiläinen, T., Kuosmanen, T. & Kumbhakar, S.C. 2008. Measuring productivity dif- ferentials - An application to milk pro- duction in Nordic countries. In: XIIth Congress of the European Association of Agricultural Economists “People, Food and Environments: Global Trends and European Strategies”. 7 p. Sjaastad, E. & Bromley, D. 1997. Indigenous Land Rights in Sub-Saharan Africa: Ap- propriation, Security and Investment Demand. World Development 25: 549– 562. Soule, M., Tegene, A. & Wiebe, K. 2000. Land tenure and the adoption of con- servation practices. American Journal of Agricultural Economics 82: 993–1005. Swinnen, J., Ciaian, P. & Kancs, A. 2009. Study on the Functioning of Land Mar- kets in the EU Member States under the Influence of Measures Applied under the Common Agricultural Policy. Centre for European Policy Studies. Turtola, E. & Ylivainio, K. 2009. Suomen ko- tieläintalouden fosforikierto – säästöpo- tentiaali maatiloilla ja aluetasolla. Maa- ja elintarviketalous 138. USDA. http://www.nass.usda.gov/census/ census97/aelos/tbl69.pdf Vermunt, J.K. & Van Dijk., L. 2001. A non- parametric random-coefficients ap- proach: the latent class regression 34 MTT SCIENCE 4 model. Multilevel Modelling Newsletter 13:6–13. Viljavuuspalvelu. 2000. Peltojen pH vuosina 1955-2005. (In Finnish). Title in English: pH levels in Finnish farmlands on 1955- 2005. In internet: http://www.viljavuus- palvelu.fi/user_files/files/Tilastoja/2007/ Happamuus_koko_maa_1955-2005.xls. Referred 05.03.2009. Yearbook of farm statistics. Maa- ja met- sätalousministeriön tietopalvelukeskus, Helsinki. Ylätalo, M. 1987. Maatalouden tuottavuus ja investoinnit. Pellervon taloudellisen tutkimuslaitoksen julkaisuja 8. MTT is publishing its research fi ndings in two series of publications: MTT Science and MTT Growth. The MTT Science series includes scientifi c presentations and abstracts from conferences arranged by MTT Agrifood Research Finland. Doctoral dissertations by MTT research scientists will also be published in this series. The topics range from agricultural and food research to environmental research in the fi eld of agriculture. MTT, FI-31600 Jokioinen, Finland. Tel. +358 3 4188 2327, email julkaisut@mtt.fi 4 Land leasing, land degradation and agricultural productivity in Finland Doctoral Dissertation Sami Myyrä MTT CREATES VITALITY THROUGH SCIENCE www.mtt.fi /julkaisut