Impact of Aging and Underemployment on Income Disparity between Agricultural and Non-Agricultural Households

Round 1

Reviewer 1 Report

This is my report on the paper "Impact of Aging and Underemployment on Income Disparity between Agricultural and Non-agricultural Households", submitted to be considered for publication in Sustainability. I think that the topic presented here is interesting and that the authors made a considerable effort to estimate the impacts of diverse policy scenarios.

More specifically, the authors aim at quantifying the expected changes in household income in South Korea when different scenarios regarding the population aging and underemployment are simulated. They base on a panel-structured dataset with a representative recent sample of Korean households and make use of a panel-data fixed effects estimator to produce their estimated impacts. Basing on these measurements, they conclude that income disparity between agricultural and non-agricultural household could be reduced by the implementation of taylored policies addressing aging and underemployment problems.

My impression is that the authors should clarify some points related to the data and the estimation methodology applied on their study. I am focusing on this empirical part since is closer to my area of expertise.

The main points I consider that should be addressed are the following:

1. From the explanation given on page #4 (lines 153 to 165 approximately) I understand that the authors opted for a fixed-effects estimator. If this is the case, equation (1) should be conveniently modified to accommodate the presence of these household-specific time-invariant components. As currently expressed, this equation only consider time dummies.
2. It seems that the authors focus on the elasticities of the regressors and not so much on the estimates of the coefficients of a linear model. I don't have a problem with that, but given the interest of the authors in these estimated elasticities I wonder to what extent the chosen linear specification of the model should be preferred to a logarithmic form. By expressing the dependent variable and the covariates in logs, they could produce estimates of the elasticities directly, without the transformation expressed in equation (4) and, perhaps more importantly, assuming that these elasticities are constant across the distribution of the covariates. In other words, they would not need to evaluate in an arbitrary point as the mean value.
3. In the specified model, the authors seem to include the joint effect of "aging" and "underemployment rate". Even when it seem obvious that this is done by means of an interaction of these two variables, I think that the manuscript would benefit of expressing this more clearly on Table 2.
4. Something I don't completely understand is why the authors state that the structure of the dataset produces an unbalanced panel (page #5, line 227). Given the annual replacement of 5% of the households in the sample, would not be enough with keeping the remaining 60% of the sample? I recognize that attrition could be an issue, but given the sample size, I wonder if this is actually problematic and having a smaller but balanced panel instead would be preferable.    
5. Some lines in the text are colored in blue. Please correct this.

Author Response

Impact of Aging and Underemployment on Income Disparity between Agricultural and Non-agricultural Households (Manuscript ID: Sustainability-1403320).

 

Response to Reviewer 1:

 

Thank you for your review comments on our manuscript. To facilitate your review of our revisions, we have first paraphrased your concern and then stated how it has been addressed.

1. From the explanation given on page #4 (lines 153 to 165 approximately) I understand that the authors opted for a fixed-effects estimator. If this is the case, equation (1) should be conveniently modified to accommodate the presence of these household-specific time-invariant components. As currently expressed, this equation only consider time dummies.

 

Following the reviewer’s suggestion, we included the time-invariant household effect term in equation (1) on page 4.

 

On page 4:

 

Consider the following longitudinal regression model,

,

(1)

where  is the per capita income of household i in year t,  is a year-specific effect,  is a household effect,  represents covariates and  is an idiosyncratic error term. In equation (1), the parameter vector,  represents covariates’ marginal effect on household i's income over time. The covariates, , include ratio of aged individuals (age over 65) for each household,  unemployment ratio, underemployment ratio, out of the labor force ratio, financial asset value, real estate value, and household head’s education level.

2. It seems that the authors focus on the elasticities of the regressors and not so much on the estimates of the coefficients of a linear model. I don't have a problem with that, but given the interest of the authors in these estimated elasticities I wonder to what extent the chosen linear specification of the model should be preferred to a logarithmic form. By expressing the dependent variable and the covariates in logs, they could produce estimates of the elasticities directly, without the transformation expressed in equation (4) and, perhaps more importantly, assuming that these elasticities are constant across the distribution of the covariates. In other words, they would not need to evaluate in an arbitrary point as the mean value.

 

Thank you for the suggestion. We initially considered the log-log functional form for the conveniences that the reviewer suggested. However, we could not use the log-log functional form because our ratio variables (e.g., aging ratio, underemployment ratio, etc.) contain many observations with zero values. These observations could have been excluded from the regression if we used the log-log functional form. We added the reviewer’s comments on page 10 of our revised manuscript for our readers.

 

On page 10:

To compare the importance of aging and household members’ employment status in determining household income with consideration of both direct and indirect (interaction) effects, we calculate elasticities of household income (HI) with respect to each of AR, UN, UE, and OL. In general, the log-log functional form would be plausible to obtain elasticities measuring the unit-free marginal effects (Kilpatrick 1973; Wellington 1991; Espey et al. 1997). However, the double log functional form could not be used in our study because our ratio variables (e.g., aging ratio, underemployment ratio, etc.) contain a large number of observations with zero values.  These observations could have been excluded if we used the double-log functional form.

 

3. In the specified model, the authors seem to include the joint effect of "aging" and "underemployment rate". Even when it seems obvious that this is done by means of an interaction of these two variables, I think that the manuscript would benefit of expressing this more clearly on Table 2.

 

We appreciate your suggestion. We revised all interaction notations in Table 2. For example, “AR and UN” becomes “AR*UN.” 

 

4. Something I don't completely understand is why the authors state that the structure of the dataset produces an unbalanced panel (page #5, line 227). Given the annual replacement of 5% of the households in the sample, would not be enough with keeping the remaining 60% of the sample? I recognize that attrition could be an issue, but given the sample size, I wonder if this is actually problematic and having a smaller but balanced panel instead would be preferable. 

   

In labor panel data, it is common that a part of survey participants drops out (due to death or immigration) from the panel over time. Without replacing the dropouts with new entrants, the panel studies using only balanced data would lose credibility of representing population behavior (Baltagi and Song 2006). Moreover, it would cause an attrition bias that would lead to biased study results (Lugtig 2014). This attrition bias could be critical especially for economic behavioral variables such as unemployment (Maluccio 2004). Therefore, we think that the unbalanced form of our current data work better for our purpose than the shortened and underrepresenting balanced data.

5. Some lines in the text are colored in blue. Please correct this.
   • Again, thank you. All blue parts are now colored in black.

 

References:

Baltagi, B.H., and S.H. Song, 2006. "Unbalanced panel data: A survey." Statistical Papers, 47 (4): 493-523.

Espey, M., Espey, J., and W.D. Shaw, 1997. "Price elasticity of residential demand for water: A meta‐analysis." Water Resources Research, 33 (6): 1369-1374.

Kilpatrick, R.W. 1973. "The income elasticity of the poverty line." The Review of Economics and Statistics, 327-332.

Lugtig, P. 2014. "Panel attrition: separating stayers, fast attriters, gradual attriters, and lurkers." Sociological Methods & Research, 43 (4): 699-723.

Maluccio, J.A. 2004. "Using quality of interview information to assess nonrandom attrition bias in developing‐country panel data." Review of Development Economics, 8 (1): 91-109.

Wellington, A.J. 1991. "Effects of the minimum wage on the employment status of youths: An update." Journal of Human Resources, 27-46.

Reviewer 2 Report

I would recommend state research questions and hypotheses explicitly.

Author Response

Impact of Aging and Underemployment on Income Disparity between Agricultural and Non-agricultural Households (Manuscript ID: Sustainability-1403320).

 

Response to Reviewer 2:

 

Thank you for your review comments on our manuscript. To facilitate your review of our revisions, we have first paraphrased your concern and then stated how it has been addressed.

1. I would recommend state research questions and hypotheses explicitly.

 

Thank you for the suggestion. We included objectives more clearly on page 2.

 

On page 2: The important question we seek to answer in this study is: are aging and underemployment major factors of household income and income disparity between agricultural and non-agricultural households? If they are, what would be the appropriate policy direction to address this problem? Although the literature provides ample evidence that aging and underemployment play a significant role in the economic condition of agricultural and non-agricultural households (e.g., Friedland and Price 2003; Loughrey and Hennessy 2014; Golub and Hayat 2015; Guo et al 2015; Bell and Blanchflower 2018; Seok et al. 2018; Du et al. 2019), it has done little to empirically answer the aforementioned question. To answer the question, we first estimate three longitudinal models for entire households, agricultural households, and non-agricultural households. Then, to examine the importance of aging and underemployment in determining household income and income disparity between the two sectors, marginal effects of aging and underemployment on household income are calculated in elasticity form using estimates from the three longitudinal regressions. Third, the income disparity is estimated using both longitudinal and cross-sectional models. Finally, the estimated disparity is simulated under five scenarios of reduced aging and underemployment to see if government policies to reduce aging and underemployment problem in agriculture could mitigate the current income disparity between agricultural and non-agricultural households. Previous studies also report that aging and employment status, including underemployment, are likely endogenous (Ham 1982; Jackle and Himmler 2010; Aiyar and Ebeke 2016). Therefore, we use a fixed-effect longitudinal model with Gaussian copula correction procedure to control the endogeneity and unobservable effects (e.g., change in government policies).

References:

Aiyar, S., and C. Ebeke, 2016. “The Impact of Workforce Aging on European Productivity.” IMF Working Papers, 16 (238): 1. https://doi.org/10.5089/9781475559729.001.

Bell, D.N., and D.G. Blanchflower, 2018. "The lack of wage growth and the falling NAIRU." National Institute Economic Review, 245: R40-R55.

Du, H., R.B. King, and P. Chi, 2019. "Income inequality is detrimental to long-term well-being: A large-scale longitudinal investigation in China." Social Science & Medicine, 232: 120-128.

Friedland, D.S., and R.H. Price. 2003. "Underemployment: Consequences for the health and well-being of workers." American Journal of Community Psychology, 32 (1-2): 33-45.

Golub, S., and F. Hayat, 2015. "Employment, unemployment, and underemployment in Africa." The Oxford Handbook of Africa and Economics, 1: 136-153.

Guo, G., Q. Wen, and J. Zhu, 2015. "The impact of aging agricultural labor population on farmland output: from the perspective of farmer preferences." Mathematical Problems in Engineering, 2015 (4): 1-7.

Ham, J.C. 1982. "Estimation of a labour supply model with censoring due to unemployment and underemployment." The Review of Economic Studies, 49 (3): 335-354.

Jäckle, R., and O. Himmler, 2010. "Health and Wages Panel data estimates considering selection and endogeneity." Journal of Human Resources, 45 (2): 364-406.

Loughrey, J., and T. Hennessy, 2014. "Hidden underemployment among Irish farm holders 2002–2011." Applied Economics, 46 (26): 3180-3192.

Seok, J., H. Moon, G. Kim, and M.R. Reed, 2018. "Is aging the important factor for sustainable agricultural development in Korea? evidence from the relationship between aging and farm technical efficiency." Sustainability, 10 (7): 2137.

 

 

 

 

Reviewer 3 Report

         The study of peasant household income and its influencing factors has always been the focus of development economics or agricultural economics. Using panel micro-survey data, the author tries to analyze the grouped effects of aging and underemployment on farmers' agricultural income and off-farm income. In general, the research topic has certain significance and logical structure is reasonable. Some suggestions are as follows: 

         (1) The marginal contribution of research is not clear. In fact, there have been too many researches on farmers' income, and the literature review of the author is not in-depth. The marginal contribution of this research cannot be seen from the research content, research methods, theoretical analysis and other aspects, which needs to be further clarified. 

         (2) Aging and underemployment are negatively correlated with income, which is obvious and can be achieved without complex econometric models. At the same time, the author did not explain the mechanism of action between the core variables in depth, which greatly reduced the theoretical analysis of this study. 

         (3) There are many factors that affect farmers' income. Why it needs to be further explained from the perspective of aging and underemployment? Some macro statistical data can be cited to further clarify the research background. 

         (4) The results should be interpreted in more detail. For example, what is the basis for the selection of fixed effects model, random effects model and mixed OLS regression?  Relevant test statistics should be disclosed.  Meanwhile, why the data of 2009, 2012 and 2015 are only given in table 1 also needs to be further explained. 

Author Response

Impact of Aging and Underemployment on Income Disparity between Agricultural and Non-agricultural Households (Manuscript ID: Sustainability-1403320).

 

Response to Reviewer 3:

 

Thank you for your review comments on our manuscript. To facilitate your review of our revisions, we have first paraphrased your concern and then stated how it has been addressed.

 

1. The marginal contribution of research is not clear. In fact, there have been too many researches on farmers' income, and the literature review of the author is not in-depth. The marginal contribution of this research cannot be seen from the research content, research methods, theoretical analysis and other aspects, which needs to be further clarified. 

 

In response to the reviewer’s comments, we added more discussions to Literature Review and stated objectives and procedures more clearly in Introduction.

 

On page 3 in Literature Review:

 

Many earlier studies find that income level is highly correlated with aging and underemployment rate. In these studies, factors affecting agricultural household income include household’s economic conditions, conditions of farmland, regional economic environments, and farm policies (Yang 1999; Dagum and Slottje 2000; Benayas et al. 2007; Sicular et al. 2007; Qian and Smyth 2008: Fisher et al. 2010; Imai and Malaeb 2016; Zhang and Posso 2019). However, only few studies discuss the potential impact of aging and underemployment on agricultural households. Some studies consider age of farm operators or number of laborers (Zhang et al. 2014; Su et al. 2018) as important factors of agricultural household productivity. Nonetheless, household-level aging or employment status (e.g., underemployment) has rarely been examined to study agricultural income and income disparity in the literature.

On page 2 in Introduction:

The important question we seek to answer in this study is: are aging and underemployment major factors of household income and income disparity between agricultural and non-agricultural households? If they are, what would be the appropriate policy direction to address this problem? Although earlier studies in the literature provide ample evidence that aging and underemployment play a significant role in the economic condition of agricultural and non-agricultural households (e.g., Friedland and Price 2003; Loughrey and Hennessy 2014; Golub and Hayat 2015; Guo et al 2015; Bell and Blanchflower 2018a; Seok et al. 2018; Du et al. 2019), it has done little to empirically answer the aforementioned question. To answer the question, we first estimate three longitudinal models for entire households, agricultural households, and non-agricultural households. Then, to examine the relative importance of aging and underemployment in determining household income and income disparity between the two sectors, marginal effects of aging and underemployment on household income are calculated in elasticity form using estimates from the three longitudinal regressions. Third, the income disparity is estimated using both longitudinal and cross-sectional models. Finally, the estimated disparity is simulated under five scenarios of reduced aging and underemployment to see if government policies to reduce aging and underemployment problem in agriculture could mitigate the current income disparity between agricultural and non-agricultural households. Previous studies also report that aging and employment status, including underemployment, are likely endogenous (Ham 1982; Jackle and Himmler 2010; Aiyar and Ebeke 2016). Therefore, we use a fixed-effect longitudinal model with Gaussian copula correction procedure to control the endogeneity and unobservable effects (e.g., change in government policies).

 

2. Aging and underemployment are negatively correlated with income, which is obvious and can be achieved without complex econometric models. At the same time, the author did not explain the mechanism of action between the core variables in depth, which greatly reduced the theoretical analysis of this study. 

 

Thank you for your comment. We discussed about construction of our model and the expected relationships between core variables in the “Methodology” section. However, in response to the reviewer’s comments, we added more discussions to page 5:

 

On page 5:

Many studies suggest that population aging has been a growing tendency, which greatly affects the employment status of aged individuals (Johnson 2012). Under this environment, aged workers are more likely to be at risk of being underemployed, especially in developed countries (Virick 2011). Findings from these studies suggest high correlation between aging and underemployment. To incorporate these findings in our analysis, interaction terms between aging and variables representing employment status (i.e., unemployment, underemployment, and out of labor force) as a part of covariates  in equation (1).

 

3. There are many factors that affect farmers' income. Why it needs to be further explained from the perspective of aging and underemployment? Some macro statistical data can be cited to further clarify the research background. 

 

We appreciate your suggestions. To address your suggestions, we included the following paragraph on page 2.

 

On page 2:

Many studies in the labor economics point out that aging and underemployment are major factors of determining wage, well-being, and productivity level of workers (e.g., Friedland and Price 2003; Bell and Blanchflower 2018a; Seok et al. 2018; Du et al. 2019). A few studies specifically argue that aging and underemployment become more prevalent and problematic in agricultural sector than non-agricultural sector, which could be two major factors affecting the income disparity between agricultural and non-agricultural households. For example, Lee et al. (2013) show that the Korea Gini index has increased from 0.330 to 0.342 between 2006 and 2011, and population aging has significant effect on the inequality index. Bell and Blanchflower (2018a, 2018b) find that for the post-Great Depression period in the U.K. and U.S., underemployment has a more significant role in wages than unemployment for all industries. In addition, Loughrey and Hennessy (2014) show that underemployment rate increased by 10% from 2002 to 2010 in the Irish agricultural sector, and the change in underemployment rate was significantly correlated with change in agricultural household income. Previous studies provide ample evidence that aging and underemployment play a significant role in the economic condition of agricultural and non-agricultural households (e.g., Friedland and Price 2003; Loughrey and Hennessy 2014; Golub and Hayat 2015; Guo et al 2015; Bell and Blanchflower 2018a; Seok et al. 2018; Du et al. 2019). However, little has done in the literature to empirically examine effects of aging and underemployment on household income and income disparity between agricultural and non-agricultural sectors.

             

4. The results should be interpreted in more detail. For example, what is the basis for the selection of fixed effects model, random effects model and mixed OLS regression?  Relevant test statistics should be disclosed.  Meanwhile, why the data of 2009, 2012 and 2015 are only given in table 1 also needs to be further explained. 

 

As suggested, we included the Hausman test statistics on page 5. 

 

On page 5:

The Hausman test rejects the random effect model in favor of fixed effect model in our study at the 1% level (Chi-square (statistic = 682.92, df = 12, p-value < 0.001).

 

We also explained why years 2009, 2012, and 2015 are presented in Table 1 on page 7.

 

On page 7:

Table 1 shows descriptive statistics of key variables used in this study for years 2009, 2012, and 2015. Given the limited space available, the descriptive statistics are presented only for the selected three years to show how the key variables change over time (a full description of data for all years used in this study is available from authors upon request).

 

References:

Aiyar, S., and C. Ebeke, 2016. “The Impact of Workforce Aging on European Productivity.” IMF Working Papers, 16 (238): 1. https://doi.org/10.5089/9781475559729.001.

Bell, D. N., and D.G. Blanchflower, 2018a. "The lack of wage growth and the falling NAIRU." National Institute Economic Review, 245: R40-R55.

---. 2018b. Underemployment in the US and Europe. National Bureau of Economic Research.

Benayas, J.R., A. Martins, J.M. Nicolau, and J.J. Schulz, 2007. "Abandonment of agricultural land: an overview of drivers and consequences." CAB Reviews: Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources, 2 (57): 1-14.

Dagum, C., and D.J. Slottje. 2000. "A new method to estimate the level and distribution of household human capital with application." Structural Change and Economic Dynamics, 11 (1-2): 67-94.

Du, H., R.B. King, and P. Chi, 2019. "Income inequality is detrimental to long-term well-being: A large-scale longitudinal investigation in China." Social Science & Medicine, 232: 120-128.

Friedland, D.S., and R.H. Price, 2003. "Underemployment: Consequences for the health and well-being of workers." American Journal of Community Psychology, 32 (1-2): 33-45.

Fisher, M., J.J. Reimer, and E.R. Carr, 2010. "Who should be interviewed in surveys of household income?" World Development, 38 (7): 966-973.

Golub, S., and F. Hayat, 2015. "Employment, unemployment, and underemployment in Africa." The Oxford Handbook of Africa and Economics, 1: 136-153.

Guo, G., Q. Wen, and J. Zhu, 2015. "The impact of aging agricultural labor population on farmland output: from the perspective of farmer preferences." Mathematical Problems in Engineering, 2015 (4): 1-7.

Ham, J.C. 1982. "Estimation of a labour supply model with censoring due to unemployment and underemployment." The Review of Economic Studies, 49 (3): 335-354.

Imai, K.S., and B. Malaeb, 2016. Asia's rural-urban disparity in the context of growing inequality. Research Institute for Economics and Business Administration, Kobe University.

Jäckle, R., and O. Himmler, 2010. "Health and Wages Panel data estimates considering selection and endogeneity." Journal of Human Resources, 45 (2): 364-406.

Johnson, R.W. 2012. "Older workers, retirement, and the great recession." New York: Russell Sage Foundation, 1-7.

Lee, H.Y., J. Kim, and B.C. Cin, 2013. "Empirical analysis on the determinants of income inequality in Korea." International Journal of Advanced Science and Technology, 53 (1): 95-109.

Loughrey, J., and T. Hennessy, 2014. "Hidden underemployment among Irish farm holders 2002–2011." Applied Economics, 46 (26): 3180-3192.

Qian, X., and R. Smyth. 2008. "Measuring regional inequality of education in China: widening coast–inland gap or widening rural–urban gap?" Journal of International Development: The Journal of the Development Studies Association, 20 (2): 132-144

Seok, J., H. Moon, G. Kim, and M. Reed, 2018. "Is aging the important factor for sustainable agricultural development in Korea? evidence from the relationship between aging and farm technical efficiency." Sustainability, 10 (7): 2137.

Su, G., H. Okahashi, and H. Chen, 2018. "Spatial pattern of farmland abandonment in Japan: Identification and determinants." Sustainability, 10 (10): 3676.

Sicular, T., Y. Ximing, B. Gustafsson, and L. Shi. 2007. "The urban–rural income gap and inequality in China." Review of Income and Wealth, 53 (1): 93-126.

Yang, D.T. 1999. "Urban-biased policies and rising income inequality in China." American Economic Review, 89 (2): 306-310.

Virick, M. 2011. "Underemployment and Older Workers." In Underemployment: Psychological, Economic, and Social Challenges, edited by Douglas C. Maynard and Daniel C. Feldman, 81-103. New York, NY: Springer New York.

Zhang, Y., X. Li, and W. Song, 2014. "Determinants of cropland abandonment at the parcel, household and village levels in mountain areas of China: A multi-level analysis." Land Use Policy, 41: 186-192.

Zhang, Q., and A. Posso, 2019. "Thinking inside the box: A closer look at financial inclusion and household income." The Journal of Development Studies, 55 (7): 1616-1631.

Round 2

Reviewer 3 Report

I have no other comments, thank you.