Behavioral economics
Habib Morovat; Ali Asghar Salem; Shayan Mohammad Sharifi
Abstract
Several factors influence the growth and development of the stock market. One of these factors is the behavior and performance (investment return) of individual investors. Individual investors are motivated to invest in the stock market for various reasons, such as long-term capital growth, dividends, ...
Read More
Several factors influence the growth and development of the stock market. One of these factors is the behavior and performance (investment return) of individual investors. Individual investors are motivated to invest in the stock market for various reasons, such as long-term capital growth, dividends, and hedging against the decline in purchasing power due to inflation. However, their performance in the market, in addition to general economic and stock market conditions, depends on individual characteristics. In this respect, the present study aimed to examine the significance of demographic, behavioral and investment-related factors on the performance of individual investors in the Tehran Stock Exchange. The demographic factors included age, gender, risk tolerance, patience level, etc. Behavioral biases included factors such as overconfidence and loss aversion, while the investment-related factors encompassed experience, skills, and the frequency of portfolio restructuring. Using a systematic sampling, the study collected the data from 240 questionnaires completed by individual investors in the Tehran Stock Exchange. The ordinal logit model was employed to analyze the data. The results showed that age, gender, and risk tolerance did not significantly affect the performance of individual investors. However, the patience level had a positive and significant effect on performance, with more patient investors achieving higher returns. Overconfidence and loss aversion were found to have a significantly negative effect on performance. Finally, investment experience and skills had a significantly positive effect on the investor performance.IntroductionTraditional finance assumed that investors make rational decisions in the stock market, particularly regarding risk-return trade-offs and utility maximization. However, psychologists have found that human behavior is not as rational as economists assume. Stock market anomalies and empirical research show that investors often act in ways that deviate from rational expectations. These anomalies can be explained by the emerging field of behavioral finance. Behavioral finance explores how psychological factors influence the actions of individuals or groups as investors, analysts, and portfolio managers. It also seeks to understand how emotions and cognitive biases affect the behavior of individual investors (Kengatharan & Kengatharan, 2014).Behavioral biases are defined as systematic errors in judgment. Recent studies have identified over 100 such biases in individual investor behavior. Some researchers refer to these biases as heuristic rules (rules of thumb), while others describe them as beliefs, judgments, or preferences. Psychological factors include heuristic rules or cognitive shortcuts related to information processing, memory errors, emotional and/or motivational influences, and social factors such as family upbringing or cultural norms (Pompian, 2021).The Tehran Stock Exchange (TSE) has been one of the most profitable markets (in terms of nominal yield) in the world in recent years. Compared to other parallel investment markets, it has achieved the highest nominal returns over the past 10 and 5 years. The present study aimed to assess the effect of individual factors on the performance of individual investors in the TSE. These factors were grouped into three general categories: economic–demographic characteristics (e.g., age, gender, education, income level, risk tolerance, and patience level), behavioral biases (e.g., loss aversion and overconfidence), and investment-related characteristics (e.g., investment skills, experience in the stock market, and the frequency of portfolio restructuring. The data was collected in 2021 and 2022. A systematic sample of 240 individual investors was surveyed both in-person and online. Multivariate regression and ordinal logit models were used for data analysis.Materials and MethodsUsing a systematic sampling, the study selected a sample of 240 investors active in the TSE. The data was collected through in-person and online surveys. Multivariate regression was applied to examine the causal relationship between the factors influencing the individual investors’ returns. Since the dependent variable was ordinal, the ordinal logit model was used to conduct the analysis.Results and DiscussionIn order to see if there is a significant relationship between explanatory and dependent variables, we used multiple regression. Since the dependent variable is an ordinal variable, the ordinal logit regression model was used to fit the model. The model fitting results are presented in the table below.Table 1. Model Estimation Results Using the Ordered Logit Model MethodProbabilityOdd ratioCoef*Variables0.3240.7533-0.2833Gender0.4040.9899-0.0101Age0.6331.07690.0760Education0.1591.18040.1659Income0.3790.9627-0.0380Risk tolerance0.0002.8739***1.0557Patience0.0130.7539**-0.2824Overconfidence0.0530.9867*-0.0136Loss aversion0.0031.4317***0.3589Experience0.0281.3268**0.2828Investment skill0.0571.2929*0.2561Stock share0.3290.9408-0.0610Portfolio restructuring0.2300.8534-0.1585Initial investment0.1903Pseudo R2***134.82LR Chi2(8)Source: research calculations based on STATA software *, **, *** Coefficients are significant at 10%, 5%, and 1% levels, respectively.All diagnostic control tests for goodness of fit were conducted, including tests for heteroscedasticity (Breusch-Pagan test), collinearity (VIF test), and parallel regressions across different categories (Brant’s test). The null hypotheses of no heteroscedasticity, no collinearity, and the parallel regression assumption were not rejected. In non-linear models such as the logit model, the parameter sign and p-value provide insights into the direction and significance of the effect of coefficients. However, interpreting results directly on the log-odds scale is often impractical. Similarly, expressing results in terms of odds ratios presents challenges, as odds ratios are frequently misunderstood. One common misconception is treating odds ratios as probabilities, which they are not. For these reasons, it is more meaningful to interpret logit models using predicted probability scale, as this approach aligns better with research questions focused on understanding how covariates affect the probability of an event occurring. Table 2 illustrates how the probability of being placed in each category of the dependent variable (investment performance) changes with variations in explanatory variables (patience, overconfidence, investment experience, and financial literacy)-assuming all other variables are held constant at their mean values. For instance, the probability that the most patient individual (patience = 5) falls into the category with the poorest investment performance (I have lost a lot) is 0.006. In contrast, the probability that the same individual belongs to the category with the best investment performance (I have made a lot of profit) is 0.285.Table 2. Final Effects of Changing Explanatory VariablesInvestment performanceI have lost a lot.I have lost.There is no change in my capital.I have made a profit.I have made a lot of profit. Explanatory variables0.275***0.338***0.278***0.102***0.006**1Patience0.117***0.24***0.385***0.24**0.016***20.044***0.118***0.34***0.451***0.046***30.016***0.048***0.198***0.616***0.122***40.006***0.018**0.87***0.604***0.285***50.057***0.145***0.367***0.293***0.035***1Overconfidence0.074***0.177***0.384***0.336***0.027***20.095***0.211***0.390***0.281***0.020***30.122***0.246***0.383***0.231***0.015***40.155***0.279***0.365***0.187***0.012**50.195**0.308***0.337***0.150***0.009*60.155***0.278***0.365***0.188***0.012**1Investment experience0.114***0.236***0.387***0.245***0.017***20.083***0.192***0.388***0.310***0.024***30.06***0.151***0.371***0.382***0.034***40.004***0.115***0.337***0.455***0.047***50.11***0.231***0.388***0.252***0.017***1Investment skills (financial literacy)0.084***0.194***0.389***0.308***0.023***20.064***0.158***0.375***0.368***0.032***30.048***0.127***0.35***0.430***0.042**40.036***0.1***0.315***0.490***0.056***5Source: Research calculations based on STATA software. *, **, *** Coefficients are significant at 10%, 5%, and 1% levels, respectively.These findings demonstrate that greater patience is associated with higher investment performance. However, the relationship between patience and the likelihood of belonging to specific investment performance categories is non-linear. For highly patient individuals, the probability of being in higher performance categories increases up to the third category and then declines. In contrast, for the variable of overconfidence, the probability of the most overconfident individual (overconfidence = 6) achieving the highest investment performance category (I have made a lot of profit) is only 0.009, which is statistically significant at the 10% level. Meanwhile, the probability of this same individual falling into the lowest performance category (I have lost a lot) is 0.195. Non-linear relationships are evident between all explanatory variables and the predicted probabilities of being in various investment performance categories. This non-linearity underscores the value of analyzing final effects over relying solely on the interpretation of coefficients and odds ratios in logit models.ConclusionThe current study aimed to examine the individual factors influencing the performance of individual investors in the TSE, incorporating new explanatory variables and a new model-fitting technique. The analysis using the ordinal logit model revealed that variables such as age, gender, and risk tolerance do not significantly impact the performance of individual investors. However, patience was found to have a positive and significant effect on investment performance, with more patient individuals achieving higher returns. Conversely, overconfidence and loss aversion exhibited negative and significant effects on performance. Additionally, investment experience and skills were shown to positively and significantly influence investors’ performance. Moreover, the findings indicated that TSE investors are affected by behavioral biases, which have a significantly negative impact on their performance, as well as on the overall efficiency and stability of the market. For example, loss aversion leads investors to hold onto losing investments for prolonged periods, reducing portfolio returns. Similarly, overconfidence causes investors to overestimate their ability to evaluate investment opportunities, often disregarding negative signals that could suggest avoiding or selling certain stocks. The study also demonstrated that investment skills (e.g., financial literacy) had a significantly positive impact on investment performance. In contrast, general literacy and unrelated educational backgrounds showed no significant effect on performance. Overall, individuals with greater investment experience and higher levels of patience tend to achieve better outcomes and earn higher returns in the TSE.
Employment
Leyla Jabari; Ali Asghar Salem
Abstract
Climate change, caused by the increase in the emission of carbon dioxide and other greenhouse gases, is one of the critical issues that mankind has faced and has created significant risks for both humans and the environment. In recent decades, many researchers have studied the factors that cause and ...
Read More
Climate change, caused by the increase in the emission of carbon dioxide and other greenhouse gases, is one of the critical issues that mankind has faced and has created significant risks for both humans and the environment. In recent decades, many researchers have studied the factors that cause and affect carbon dioxide and their control. Among the factors affecting the emission of carbon dioxide, we can mention the structural labor change, which can play an important role in increasing the emission of carbon dioxide through the increase of industrial activities and economic growth. Therefore, in the present study, the effect of structural labor change on carbon dioxide emissions in Iran’s provinces was investigated using the Quantile regression with non-additive fixed effects presented by Powell (2016). The results show that increasing labor transfer from the agricultural sector to other economic sectors, including services and industry, increases carbon dioxide emissions. Additionally, indirectly, the structural labor change index has a positive and significant effect on carbon dioxide emissions in Iran’s provinces. The study also confirmed an inverse N relationship between carbon dioxide emissions and economic growth. The coefficients obtained for income inequality are negative and significant, while those for per capita energy consumption, industrialization, and urbanization are positive and significant. IntroductionSince the early 1990s, the emission of carbon dioxide and other greenhouse gases has increased in most countries, aligning with economic growth. This has given rise to numerous challenges for humanity, inflicting detrimental effects on ecosystems across various parts of the world. The increase in carbon dioxide emissions over the past two decades has prompted researchers to delve into the factors influencing such emissions and their control. One significant factor influencing carbon dioxide emissions is the transfer of labor from the agricultural sector to other sectors. This transition is recognized as a hallmark of economic development, commonly referred to as a structural labor change in the field of development economics. Though most economic theories view the labor transfer as an indicator of socio-economic progress, this phenomenon also has disadvantages that can result in abnormal consequences affecting culture, the environment, society, and economy. Shao et al. (2021) and Yang et al. (2021) highlight it as a pivotal factor influencing carbon dioxide emissions and environmental degradation. Understanding the impact of this phenomenon on carbon dioxide emissions is crucial for formulating policies aimed at regulating the emitted carbon dioxide levels. In Iran, the transfer of labor from the agricultural sector to other economic sectors has risen, driven by diverse motives and concurrent with the expansion of urbanization and industrialization. This shift may entail numerous environmental challenges. Long-term statistics reveal that since 1956, the agricultural sector has lost its superiority, while the industrial and service sectors have experienced an increase in the number of workers. The disparity between the industry and services sectors compared to agriculture has widened (Mohinizadeh et al., 2019). However, in Iran, the impact of structural labor change on carbon dioxide emissions has not received significant scholarly attention. In this respect, the present research aimed to explore the nonlinear effects of structural labor change across 31 provinces in Iran during 2010–2020. The study first calculated the carbon dioxide emissions in each province. Subsequently, the analysis focused on the impact of structural labor change, particularly the transfer of labor from the agricultural sector to other economic sectors, on carbon dioxide emissions in the provinces. Materials and MethodsThe study adopted the experimental model proposed by Liu et al. (2019) and Yang et al. (2021), utilizing the subform presented in Equation (1). (1) Equation (1) defines the following variables: lnCO_2 represents the logarithm of carbon dioxide emissions per capita; lnGDP signifies the logarithm of real GDP; ln2GDP denotes the square of the logarithm of real GDP; ln3GDP represents the cube of the logarithm of real GDP; lnRatioagr indicates the logarithm of structural labor change; lnGini is the logarithm of income inequality; lnUrb denotes the logarithm of urbanization; lnIndst is the logarithm of industrialization; and lnEC stands for the logarithm of energy consumption. Furthermore, lnRatioagr×GDP represents the logarithm of the interaction term between structural labor change and real GDP. This variable was incorporated into the model due to the indirect impact of structural labor change on carbon dioxide emissions. In addition to the variable of structural labor change, the study examined the effect of other explanatory variables on carbon dioxide emissions. These variables are summarized in Table 1. Table 1. Introduction of explanatory variablesSourceDescriptionVariableStatistical Center of IranThe ratio represents the percentage of the working labor force in the agricultural sector compared to the total working population. A higher percentage indicates less change in the employment structure, while a lower percentage signifies more pronounced structural changes in the labor force.Structural labor changeEnergy balanceTotal energy consumption per capita, encompassing natural gas, kerosene, fuel oil, and gasoline (thousand liters).Energy consumptionStatistical Center of IranThe ratio of the added value of the industrial sector to the GDP (million rials)IndustrializationMinistry of Economic Affairs and FinanceReal GDP (million rials).Economic growthStatistical Center of IranGini coefficient of total consumption expenditure of urban and rural households in each province, weighted by population (percentage)Income inequalityStatistical Center of IranThe ratio of the urban population in each province to the total population of the province (percentage)Urbanization Results and DiscussionFocusing on the transfer of labor from rural and agricultural areas to urban and industrial or service centers, the present study investigated the impact of this labor transfer on carbon dioxide emissions across 31 provinces in Iran during 2010–2020. First, the carbon dioxide emissions for each province were calculated. Then, the study introduced a model based on quantile regression with nonadditive fixed effects at varying quantile levels. The primary rationale behind employing this regression technique was to offer a detailed and comprehensive analysis of the model’s response variable. This approach allows for intervention not only at the center of gravity of data but also at all levels of the distribution particularly the extremes avoiding the issues associated with assumptions such as ordinary regression, heterogeneity of variance, and the potential impact of outlier data on coefficient estimations. Consequently, the panel quantiles were used to estimate the regression model, and the results are presented in Tables 2 and 3.Table Table 2. Estimation results ation resultsvariables / (τ)5040302010 -48.59***-30.69***29.14***-24.32***-24.46*** 3.13***1.94***1.84***1.52***1.56*** -0.067***-0.041***-0.039***-0.032***-0.033*** -0.622***-0.592***-0.508***-0.758***-0.525*** -0.161-0.068***-0.120***-0.117***-0.202*** 0.0520.722***0.996***1.089***0.918*** 0.143***0.123***0.096***0.103***0.076*** 0.614***0.684***0.646***0.662***0.719*** 0.038***0.046***0.044***0.059***0.042***Source: Research resultsTble Table 3. Estimation results ation resultsvariables / (τ)90807060 -154.60***-73.48***-41.63***-46.65***ln2GDP9.99***4.73***2.96***30.9*** -0.214***-0.101***-0.058***-0.068***Ratioagri-1.99**-0.221***0.0070.612 -0.144-0.257***-0.017-0.046***lnUrb0.340***0.0360.184***0.396*** 0.106***0.130***0.135***0.128*** 0.645***0.724***0.671***0.586*** 0.126**0.015**0.0007-0.039Note: ***, ** and * represent the significance level of 1, 5 and 10%, respectively.Source: Research resultsIncreasing the proportion of the working population in the agricultural sector relative to other sectors or minimizing changes in the labor structure, except between the 60th and 70th percentiles, leads to a reduction in carbon dioxide emissions. As a result, the structural labor change exerts a direct and significant impact on the levels of carbon dioxide emissions across Iran’s provinces. As changes in the labor structure intensify, the agricultural sector might resort to machinery to compensate for the workforce reduction, maintaining production and moving towards capitalization that, in turn, amplify energy consumption and carbon dioxide emissions. Furthermore, the transition from rural areas and agricultural hubs to urban and industrial centers can increase income, thereby contributing to an increase in carbon dioxide emissions. The study also examined the indirect impact of structural labor change on carbon dioxide emissions through the economic growth channel. According to the estimation results, the coefficient for the interaction term of structural labor change and economic growth is positive and statistically significant in all quantiles, except the 60th and 70th percentiles. As noted by Yang et al. (2021), the increased transfer of labor from the agricultural sector to other sectors, particularly industry, during the course of economic development can indirectly boost economic growth and carbon dioxide emissions. The labor transfer increases as the scale and GDP rise, and there is an expansion in fossil fuel consumption accompanying economic growth, leading to a subsequent increase in carbon dioxide emissions in the provinces of Iran. The study validated two direct and indirect effects of structural labor change on carbon dioxide emissions in Iran’s provinces. In both scenarios, structural labor change contributed to an increase in carbon dioxide emissions. The first effect stems from the increasing use of machinery to compensate for the labor force depleted from the agricultural sector, leading to increased energy consumption and subsequent carbon dioxide emissions. The second effect can be explained with an eye to the increased economic growth and GDP resulting from the structural labor change, as discussed in the Lewis model. ConclusionThe study examined both the direct and indirect effects of structural labor change, in conjunction with other socio-economic variables, using a nonlinear method. The data was gathered from 31 provinces of Iran spanning from 2010–2020, and the study used a quantile regression with nonadditive fixed effects. The variable denoting labor transfer from the agricultural sector to other sectors was used as the ratio of the working population in the agricultural sector to the total working population, serving as the index for structural labor change. The findings revealed that structural labor change has a direct effect on carbon dioxide emissions. Furthermore, concerning indirect effects, it can be affirmed that the index has a positive and significant effect on the dependent variable through the indirect channel of economic growth. Considering the positive effect of labor transfer and its negative impact on carbon dioxide emissions and environmental degradation, it is recommended to adopt measures to control and regulate the labor transfer. Specifically, strategies should be devised to increase the income of workers in the agricultural sector, aiming to establish an equitable wage balance relative to other sectors. Moreover, provincial authorities should prioritize initiatives that increase the real added value in agriculture, with a focus on expanding industries associated with agricultural production, such as transformative and complementary sectors.
Neda Bayat; Ali Asghar Salem
Abstract
The critical situation of water and its demand growth in Iran as well as the destruction and improper exploitation of groundwater resources with drought continuity have caused the management of water demand to become an important concern in the country's policies in all sectors, including the household ...
Read More
The critical situation of water and its demand growth in Iran as well as the destruction and improper exploitation of groundwater resources with drought continuity have caused the management of water demand to become an important concern in the country's policies in all sectors, including the household sector. Without identifying the influential factors and their importance in the pattern of household water consumption, planning and implementation of effective policies are impossible. This study identifies socio-economic factors affecting water demand and determine their degree of importance in the pattern of household water consumption. Due to the impossibility of specifying the accurate mathematical relationships between these factors and water consumption, this study used the random forest algorithm to determine the most important factors. Also, because of the differences in the lifestyle and cultures of rural and urban households, these two groups have been studied separately. Finally, according to the obtained results, 17 important factors were determined in three different levels of influence. The variables of household income, house size, and age of the household’s head were identified as the three most important quantitative variables affecting both urban and rural household consumption, respectively. Among the qualitative variables, the use of evaporative cooler was recognized as the most important effective variable. Other factors were ranked in order of effectiveness.
Ali Asghar Salem
Abstract
In this study, in order to estimate the Engel curve and income elasticity of Information and communications technology (ICT) for residential urban households in Iran, the Working-Leser equations system functional form has been used considering economic and social characteristics of the households. This ...
Read More
In this study, in order to estimate the Engel curve and income elasticity of Information and communications technology (ICT) for residential urban households in Iran, the Working-Leser equations system functional form has been used considering economic and social characteristics of the households. This model is estimated using Seemingly Unrelated Regressions (SUR) analysis and cross section data on nearly 19 thousands urban households for the year 2015. The results show that income elasticity of ICT for low, middle, and high-income households is 1.22, 1.12 and 0.8 respectively. These results show that ICT is a luxury good for the whole society. However, this commodity group is considered as a necessary good for the wealthy households. Also increasing the level of education leads to an increase in using of information and communication technology products. One percent increase in the years of education, leads to 0.06 percent increase in demand of information and communication technology products.
Teimour Mohammadi; Ali Asghar Salem; Fatemeh Mir Mohammad Ali Tajrishi
Abstract
Equivalence scale is an important concept in household welfare debates wich plays an important role in the measurement of poverty and inequality. Equivalence scale is an index that converts household's expenditures into comparable values. In this research, equivalence scale in terms of the relative cost ...
Read More
Equivalence scale is an important concept in household welfare debates wich plays an important role in the measurement of poverty and inequality. Equivalence scale is an index that converts household's expenditures into comparable values. In this research, equivalence scale in terms of the relative cost of a child was estimated using Price scaling with a Quadratic Almost Ideal Demand System. The estimation method is nonlinear seemingly unrelated regressions and the estimation period is 2008-2012. Results indicate that one child costs about 15 percent of an adult in rural households and the quadratic expenditure effects is highly significant. It is concluded that the general equivalence scale, varies with price. Household's equivalence scales with different demographic characteristics is used to calculate equivalent income in this period in order to compare welfare, poverty and income inequality across rural households.
