The report published by EUFCI regarding the countries in in Danube-Carpathian Region indicates that the local residents and the poor people living around the forest areas are the main actors of the forest crimes in the region (Schlingemann et al., 2021). Similarly, Şen and Ünal (2011) reports that the main causes of the forest crimes in Turkey are the people living in rural areas or villages near forests. Additionally, they state that the most important reason for forest crimes in Turkey is the economic reason. In another study performed in Ilgaz province of Turkey, Ünal et al. (2021) report that the forest crimes are directly correlated with low levels of income, lack of awareness of laws, low penalties, and low education levels. Moreover, in a study performed in Black Sea region of Turkey, Durkaya et al. (2020) found that the income and education levels of the people living in the forest villages had a direct effect on the forest crimes committed.
The studies show that one of the mexamined, it is seen that the logs of coniferous and non-coniferous wood productions in 2020 are almost aain causes of the forest crimes in Turkey is the low level of incomes of the people living in the villages near the forest areas. In Turkey, the state supports the forest villagers in forms of financial supports and cooperatiy, the Gross Domestic Producve credits. Thus, to investigate the relations between the financial supports provided to forest villagers and forest crimes, Turkey is an appropriate area of study.
Turkish forests are under the control of the Orman Genel Müdürlüğü – General Directorate of Forestry (GDF), which is a state agency. Turkey has a forest area of 22,993,000 ha as of 2020 and the forest areas cover 29.4% of the country area. There are also private forests, which are less than two thousandths of the whole forest area (GDF, 2021a). The forest assets of Turkey for the year 2020 are demonstrated in Fig. 1 (GDF, 2021b).
As seen in Fig. 1, in Turkey, the GDF is divided into 28 regional directorates of forestry as of the year 2020, which are Ankara, Antalya, Artvin, Balıkesir, Bolu, Bursa, Çanakkale, Denizli, Elazığ, Erzurum, Eskişehir, Giresun, Isparta, İstanbul, İzmir, Kahramanmaraş, Kastamonu, Kayseri, Konya, Kütahya, Mersin, Muğla, Sakarya, Şanlıurfa, Trabzon, and Zonguldak directorates. The distribution of the forest assets with respect to the regional directorates of forestry, which are presented in also Fig. 1, are demonstrated in Fig. 2, based on the data published by GDF (2021a) for the year 2020.
Figure 2 shows that, in Turkey, Amasya, Elazığ, Şanlıurfa, Kastamonu, Muğla, Antalya, and İzmir are the first eight regions having the largest forest areas, which are more than one million ha. On the other hand, Giresun, Erzurum, Çanakkale, Artvin, and Sakarya are the last five regions having the least forest areas, which are less than 600,000 ha. The most common tree species observed in Turkish forests are oak (Quercus) (29.42%), Turkish pine (Pinus brutia) (22.74%), and black pine (Pinus nigra) (18.31%). The annual amounts of wood in the rough production (m3) in Turkey (GDF 2021d), between the years 2010 and 2020, are given in Table 1.
Table 1
Yearly wood production of Turkey in 2010–2020.
Years | Logs of Coniferous Wood (m3) | Logs of Non-Coniferous Wood (m3) (except tropical wood) | Fuel Wood (m3) |
2010 | 9,501,980 | 3,066,539 | 5,395,779 |
2011 | 10,440,865 | 3,141,597 | 5,083,576 |
2012 | 10,744,778 | 3,679,587 | 4,824,506 |
2013 | 10,848,147 | 2,819,840 | 4,486,277 |
2014 | 11,307,865 | 3,615,344 | 3,943,496 |
2015 | 12,807,215 | 3,830,383 | 3,767,240 |
2016 | 12,715,352 | 4,294,646 | 3,657,801 |
2017 | 11,486,044 | 4,035,579 | 3,269,735 |
2018 | 13,918,115 | 5,162,022 | 3,667,841 |
2019 | 16,252,761 | 5,860,487 | 4,192,349 |
2020 | 18,087,054 | 6,664,012 | 4,047,510 |
When Table 1 is examined, it is seen that the logs of coniferous and non-coniferous wood productions in 2020 are almost as twice as the ones in 2010. The fuel wood production, on the other hand, mostly has a decreasing trend, except for the slight increases in 2019 and 2020. |
In Turkey, the Gross Domestic Product (GDP) grew by 4.5% and the industrial sector by 6.2% in the first quarter of 2020. However, due to the negative effects of the Covid-19 pandemic, the GDP contracted by 9.9% and the industrial sector by 16.5% in the second quarter of 2020. The GDP of Turkey as of 2020 is $719.955 billion (World Bank, 2020). Although 29.4% of Turkey is covered by forests, income from the forest products has 3% contribution to the government treasure. (GDF, 2021d). According to the data published by Sosyal Güvenlik Kurumu – Social Security Institution (SSI) of Turkey, there are 34,579 workers employed in forest-based industries in 2020 (SSI, 2021).
2.1. Forest crimes and forestry financial supports in Turkey
Forest crimes, in general, can be defined as any action harming forest assets or their future and prohibited by laws to protect forests. A report by World Bank (2006) describes the forest crimes as illicit activities such as illegal logging illegal occupation of forestland, woodlands arson, wildlife poaching, and encroachment on forests (both on public and private ones). The report also states that the corruption caused by forest crimes all over the world is particularly troubling in developing countries. Although the report addresses the weak governance and subsequent poor law enforcement as the main cause of the forest crimes in the world, it also suggests that poverty reduction approaches targeted at forest-dependent populations committing forest crimes are also needed. In another study by Kishor and Belle (2004), which also supports the improved governance solution to reduce forest crimes, international trade in protected species, logging outside concession boundaries or in protected areas, underrating and misclassifying species, timber smuggling, transfer pricing in timber trade, and timber processing without a license are also considered as forest crimes. Contreras-Hermosilla (2002) provides a detailed list and descriptions of various forest crime types.
In Turkey, forest crimes and the corresponding punishments are defined and regulated by Turkish Forestry Law (1956) numbered 6831, which was published in the Official Gazette in Turkey on September 8, 1956. According to Turkish Forestry Law (1956), Article 4, there are three types of accepted forest ownership as state forest, forest belonging to public legal entities, and private forest. State forests are owned and controlled by the GDF as well as processing and manufacturing of all kinds of forest products, as mentioned in Article 89 of Turkish Forestry Law (1956). However, Article 6 states that all forest owned by the other parties than the State are still subject to the control of GDF.
Turkish Forestry Law (1956) provides detailed descriptions of many kinds of forest crimes with the corresponding prohibitions, punishments, and fines. In this study, there are six types of forest crimes, which were taken into consideration. These crimes are illegal logging of trees, illegally transferring of the forest products, illegally expanding the lands into the forests, illegal occupation of the forestlands, illegally expended trees, and illegal pasturages in the state forests.
The term illegal logging was used in this study in the context of harvesting timber in contravention of the related laws and regulations. Turkish Forestry Law (1956) defines the illegal logging crime in Article 14 as
A) “To cut or uproot grown or planted seedlings to damage plantation areas, to choke or wound trees, to cut their branches and tops or to get produce wooden tiles from the trees.”
B) “To cut old or young trees or to uproot them or to get tar or bark or resinous wood from them, to cut leaning or overthrown trees or to take or uproot them on produce coal from them.”
The crime of illegally transferring of the forest products is described in Article 108 as “anyone who transports, saws, works, accepts, sells, buys, or keeps illegally harvested or collected forest products is punished”. The term forest product refers to all timber and non-timber products that can be obtained from forests. Additionally, Article 42 states “transportations within the forest are realized in routes determined by the forest management. The transportation permits should always be carried and exposed to related personnel when requested.” Thus, according to Article 100, “transporters of any products without marking on them or without transportation permit document (against Article 41) are punished according to Article 108.”
The crime of illegally expanding lands refers to the crime of encroachment on both public and private forests as described by World Bank (2006). This type of crime is committed by expanding private lands (usually farming lands) into the forests by either burning or cutting down the trees, trespassing the forest border line. The illegal occupation crime, however, refers to any kind of building or establishment built on forestland by burning or making use the empty places through invasion, as described in Article 17 of Turkish Forestry Law (1956), which brings regulations for both of illegal expanding and illegal occupation crimes.
As well as the illegal logging crime, the crime of unlawful expense of the trees is also regarded as a crime by Article 14, which refers to using any kinds of products made from illegally obtained timber from forests, for any purpose like producing wooden tiles or coal. Moreover, Article 108 states, “Anyone who transports, saws, works, accepts, sells, buys, or keeps illegally harvested or collected forest products is punished.”
The crime of illegal pasturing is defined by Article 19 as “the access of any kind of domestic animal to forest is prohibited. The forest administration only allows grazing for animals suffering from malnutrition in drought regions.” Additionally, it also states, “this permission can be given under the terms and conditions of a given period, for the defined animal species and areas, and with the condition that no damage should be given to the forest.” Similarly, Article 21 states, “the grazing of herds on the State forest lands should be done according to the plans and permission of the forest administration.”
Logging from forests depends on strict regulations in Turkey. Turkish Forestry Law (1956) gives some rights to the Turkish citizens that are eligible to be defined as forest villagers. Forest villager documents are given to people who have been residing in a forest village in Turkey for at least one year. The list of the forest villages is determined by Tarım ve Orman Bakanlığı – Ministry of Agriculture and Forestry (MAF). Forest villagers are permitted to obtain timber and non-timber products from forests condition to necessary permissions. Article 37 states “except logs, poles, mine props, industrial wood, paper wood, fuel wood, fiber wood, stick resin, resinous wood, boxwood, storax included in the annual production program of the State, all other kinds of forest products and residues are allowed to be utilized in determined locations and periods, giving priority to forest villages, development cooperatives, or to neighboring villagers or workers as with the payment of tariff prices.” While getting these permissions, forest villagers have priority in using the forests next to their villages as mentioned in Article 40. It is also necessary to get permission for hunting in forests in Turkey according to Article 80 of Turkish forestry Law (1956). The article states, “the forest officers are authorized to detain the hunted animals and vehicles of individuals hunting in forests, forest lakes and ponds without hunting license and permission obtained from forest administration.” Elvan (2014) provides a detailed examination and explanation of the forest crime types in Turkey within the framework of criminal law.
Turkey provides monetary aids to its forest villagers in forms of individual financial supports or cooperative credits. The financial supports are provided to the forestry villagers within the frame of the law on Supporting the Development of the Forest Villagers numbered 2924. The individual financial supports are provided for the purposes like roof covering, electrical internal installations, heating – cooking, beekeeping, breeding cows and sheep – dairy farming, production mechanization, carpet and rug weaving, and facility acquisition (GDF, 2021c).
2.2. Data and variables
The data consist of the observations belonging to the dependent variables illegally logged trees, illegally transferred forest products, illegally expanded lands, illegally occupied lands, illegally expended trees, illegal pasturage numbers, and the independent variable financial supports belonging to the years 1997–2020, which was recorded and published by the GDF. The data were derived from the website of the GDF (2021c). The definitions and explanations regarding the variables used in the study are provided in Table 2.
Table 2
The variables used in the study.
| Variables | Definitions |
The dependent Variables (Forest Crimes) | Illegally logged trees (m3) | The yearly amounts of the illegally logged trees that were seized |
Illegally transferred forestry products (m3) | The yearly amounts of the forest products that were seized during their illegally transferring |
Illegally expanded lands (ha) | The yearly amounts of illegally expanded private lands into forests through encroachment |
Illegally occupied lands (ha) | The yearly amounts of illegally occupied lands in forests by building structures for settlement or business |
Illegally expended trees (m3) | The yearly amounts of seized illegally obtained trees that are used to produce wooden items |
Illegal pasturage numbers | The yearly numbers of the detected illegal pasturages |
The independent variable | Financial supports (Turkish Lira – TL) | The yearly amounts of the individual financial support (with deflators) provided to the forest villagers in Turkey by the state |
It would be useful to draw the time series plots of the variables to observe their behaviors in the years 1997–2020. Thus, the amounts of the individual financial support that were paid to forest villagers by the government and the amounts of the illegally logged trees, illegally transferred forest products, illegally expanded lands, illegally occupied lands, illegally expended trees, and illegal pasturage numbers in Turkey for the years 1997–2020 are demonstrated in Fig. 3. Because of the different scales of the variables in the data, the y-axis consists of the logarithmic scores of the values.
When Fig. 3 is examined, in general, it can be said that the individual financial supports follow an increasing pattern except some decreases in certain years such as 2001, 2008, 2014, and 2018. The probable reason for the decreases in the financial supports provided by the state in 2001, 2008, and 2018 years is the economic crises experienced in Turkey. The crisis in 2008 was actually a global crisis also affecting Turkey. In 2014, however there was a dramatic loss in the exchange rate of Turkish Lira (TL) against the US Dollar (USD), which caused a serious decrease in the purchasing power of the TL. Thus, less amount of TL was put into circulation, to prevent it from losing its value more. In 2005, 2011, 2013, and 2016 years, however, there are peak values.
In general, the crime of illegal logging displays a decreasing pattern between 1997 and 2008, except for the certain peak in 2003 and the little peak in 2007. Moreover, it stays stable between 2008 and 2013. However, after 2013, there is an extraordinary increase in the illegally logging crime level reaching its peak in 2014. The most important factor for the extraordinary increase in 2014 is that there was a considerable decrease in the value of TL against the USD in 2014, which is followed by a sharp increase in the interest rates. This situation caused a serious decrease in the purchasing power of the Turkish citizens. After 2016, however, the crime level seems to have an increasing trend. It is very notable that after the peak value of the financial supports in 2013, there is a sharp decrease reaching its trough in 2014, when the level of the illegally logged trees demonstrates an extremely high peak. The increase in the year of 2020 is notable, this increase can be explained by the conditions due to the Covid-19 pandemic. One possible reason for this extraordinary increase can be given as the curfews preventing people from working and causing a dramatic decrease in the incomes. On the other hand, the curfews also caused less security controls in the forests, which caused criminals to act more freely than usual (Lang et al, 2021); (GDF, 2021d). It can also be seen that the amounts of the illegally transferred forest products has some definite peak years like 2005, 2014, and 2017. It is possible to say that the amounts of the illegally transferred forest products visibly fluctuate between the years 1997 and 2007. It stays comparatively steady with a slightly decreasing trend between 2007 and 2011. However, there are again remarkable fluctuations with two noticable peaks in 2014 and 2017. After 2017, although there is a slight decrease until 2018, it again increases after this year. Moreover, it is also possible to see that the amounts of the illegally expanded lands fluctuate with many peaks and troughs. Still, it is possible to say that after 1997 the crime level reaches its lowest point in 2013. After this year, however, it has an increasing trend until 2018 with two slight decreases. Finally, it displays a decreasing pattern after 2019. At the same time, it is apparent that the illegally occupied lands have three extraordinary peaks in the years 2000 and 2008. The first peak can be explained by the economic crisis that Turkey experienced in 2000. In 2008, however, there was a global crisis also affecting Turkey, which caused the second greatest increase in the crime level. After the year 2017, the crime levels seem to be increasing. It is possible to comment that the illegal expense of trees crime levels have a decreasing trend except for the three small peaks in 2001, 2006, and 2009. The levels seem to be almost steady between the years 2010 and 2018, with a slight increase afterwards. When Fig. 3 is examined, it can also be seen that the numbers of illegal pasturages have a fluctuating pattern between the years 1997 and 2007. However, after 2017, the levels demonstrate a decreasing pattern until 2019. The total numbers of the forest crimes seen in the regional directorates of forest, in the year of 2020, are presented in Fig. 4.
Figure 4 shows that the highest numbers of forest crimes are seen in Amasya, Sakarya, Adana, Kahramanmaraş, Antalya, and İzmir regions, which have forest crime cases over one thousand. However, the crime levels are seen the least in Çanakkale, Kütahya, and Ankara regions, which have cases less than 250. It can be expected that that the level of crimes will be higher in the regions having larger forest areas. However, when Fig. 2 and Fig. 4 are examined together, this expectation appears to be not realistic. For example, although Amasya region has the largest forest area and the highest level of crimes, Elazığ, for example, has the second largest forest area but considerably low level of crimes. Similarly, Sakarya has a smaller forest area compared to the other regions, but the level of crimes is remarkably high in this region.
As far as the reliability of the data is concerned, apparently, the GDF provides data about various forest crimes including illegal logging, which are presented in Table 2. Thus, it is evident that there is a certain amount of illegal logging in Turkey. However, in some international studies, such as Li et al (2008), it is reported that the estimated share of the illegally logged industrial roundwood in Turkey is 0% as of 2004. Additionally, a report by the United Nations Economic Commission for Europe (UNECE) published in 2006 notes that Turkish forest law enforcement, governess, guarding and controlling system against forest crimes are strong and strict for long time, thus, illegal logging and associated forest crimes are not at high levels. Additionally, it also reports the rates of illegal logging for commercial use is quite low and not a significant issue to international trade (UNECE, 2006).
2.3. Correlation coefficient
Correlation coefficient measures the degree and the direction of the linear relation between two variables. A significant correlation coefficient also indicates a dependency relation between the variables for which it is calculated. There are various measures to calculate correlation coefficient such as Pearson and Spearman correlation coefficients. Pearson correlation coefficient is a parametric method, while Spearman correlation coefficient is a nonparametric measure of correlation. Pearson correlation coefficient requires some assumptions before it is calculated. These assumptions are linearity, continuous-level variables, homoscedasticity, normality, absence of outliers and independence.
To test whether the normality assumption required by Pearson correlation coefficient holds for the variables employed in the study, Table 3 presents the variables having a normal distribution, and the ones not normally distributed.
Table 3
Normality test results of the variables (\(\alpha =0.05\)).
Variables | Kolmogorov-Smirnov Statistic | P-value | Normally Distributed |
Illegally logged trees | 0.174 | 0.058 | Yes |
Illegally transferred forest products | 0.171 | 0.071 | Yes |
Illegally expanded lands | 0.147 | 0.150 | Yes |
Illegally occupied lands | 0.248 | 0.010 | No |
Illegally expended trees | 0.248 | 0.010 | No |
Illegal pasturage numbers | 0.140 | 0.150 | Yes |
Financial supports | 0.107 | 0.150 | Yes |
It is apparent in Table 3 that not every variable has a normal distribution, such as the variables illegally occupied lands and illegally expended trees. To check another assumption of Pearson correlation coefficient of absence of outliers, Table 4 demonstrates the outlier analysis results of the data. The outlier analysis was performed based on a nonparametric approach, which is interquartile range, as a common measure for all the variables; as it has already been shown that there are variables in the data set having non-normal distributions.
Table 4
Outlier analysis results of the variables.
Variables | N | Q1 | Q3 | Interquartile Range | Outlier – Year |
Illegally logged trees | 24 | 19,174.8 | 45,316.3 | 26,141.5 | None |
Illegally transferred forest products | 24 | 883.75 | 1,897.5 | 1,013.75 | None |
Illegally expanded lands | 24 | 967.925 | 1,368.75 | 400.825 | 2,270.60 ha – (2020) |
Illegally occupied lands | 24 | 1,025.93 | 1,613.63 | 587.7 | 3,597.20 ha – (2000) 2,837.80 ha – (2008) |
Illegally expended trees | 24 | 215.25 | 1,580.5 | 1,365.25 | 4,997 ha – (1997) 4,162 ha – (1998) |
Illegal pasturage numbers | 24 | 1,162.25 | 3,629 | 2,466.25 | None |
Financial supports | 24 | 65,139,751 | 201,412,827 | 136,273,076 | None |
Table 4 shows that the variables illegally expanded lands, illegally occupied lands, and illegally expended trees have outlier values. Additionally, to check the homoscedasticity assumption before using Pearson correlation coefficient, Table 5 presents the homoscedasticity test results between the independent variable financial supports and the dependent variables.
Table 5
Homoscedasticity test results of the variables (\(\alpha =0.05\)).
Dependent Variables | Independent Variable | Bonett’s Statistic | P-value | Homoscedasticity |
Illegally logged trees | Financial supports | 992.11 | 0 | No |
Illegally transferred forest products | 1,938.10 | 0 | No |
Illegally expanded lands | 2,125.81 | 0 | No |
Illegally occupied lands | 1,944.91 | 0 | No |
Illegally expended trees | 1,702.83 | 0 | No |
Illegal pasturage numbers | 1,616.81 | 0 | No |
When Table 5 is examined, it is seen that there is no homoscedasticity between the independent variable financial supports and any dependent variable.
It is evident that, the assumptions of normality and absence of outliers failed for some variables. Moreover, homoscedasticity assumption failed for all the variables. Furthermore, when a correlation coefficient is to be calculated between two time series, Pearson correlation coefficient cannot be used directly, as it is appropriate for independent data. However, time series data is usually dependent on time. Thus, these results indicate that Pearson correlation coefficient is not an appropriate measure to use for the variables employed in the study.
Spearman correlation coefficient, however, is a nonparametric method, which does not require any normal distribution or the other assumptions required by Pearson correlation coefficient except the linearity assumption. Thus, it can be an alternative to Pearson correlation coefficient, when its assumptions are not met. Therefore, in this study, Spearman correlation coefficient was adopted to analyze the relations between forestry financial supports and the forest crimes listed in Table 2. While calculating Spearman correlation (\({r}_{s})\) for two variables \(X\) and \(Y\), firstly they are converted to ranks as \(R\left(X\right)\) and \(R\left(Y\right)\). Then, Pearson correlation (\(\rho )\) formula is used to calculate the correlation between the ranked variables. Spearman correlation coefficient is calculated as follows.
$${{r}_{s}=\rho }_{R\left(X\right)R\left(Y\right)}=\frac{Cov[R\left(X\right), R\left(Y\right)]}{{\sigma }_{R\left(X\right)}{\sigma }_{R\left(Y\right)}}$$
1
where \({r}_{s}\) denotes Spearman correlation coefficient, \(Cov[R\left(X\right), R\left(Y\right)]\) is the covariance of the ranked variables, \({\sigma }_{R\left(X\right)}\) and \({\sigma }_{R\left(Y\right)}\) are the standard deviations of the ranked variables \(R\left(X\right)\) and \(R\left(Y\right)\) respectively. Just like Pearson correlation coefficient, Spearman correlation coefficient also varies between − 1 and + 1.