Chapter 7. Measuring the costs of well-being deprivation (Module 5)
The OECD well-being approach emphasises the relational aspects of well-being and development, recognising that development is not just about living better but about living together in a better way. In this respect, experiencing deprivation in one or more well-being dimensions comes with costs to the individual, society and the economy as a whole. This module presents a framework to estimate society and individual opportunity costs due to youth deprivations in well-being.
Calculating the costs of well-being deprivation is important for several reasons, foremost among which is that it can help make the case for investing in youth policies and programmes. First, not knowing the true costs of youth deprivation can lead to underinvestment in youth. Creating awareness of the true costs of well-being deprivation could drive policy makers to give a higher importance to youth inclusion and well-being and drive young people to reconsider some of their actions (World Bank, 2008).
Second, recognising the costs of well-being deprivation has major implications for cost-benefit analyses of youth investments. Appraisals of social programmes tend to be challenging, as the expected benefits often lack a price tag. For example, programmes aiming at keeping young people in school do not have a direct monetary benefit to balance the costs. A better understanding of the cost-savings potential of investments in youth can put a price tag on the benefits of social programmes and encourage policy makers to revisit their policy choices.
Third, from a public finance perspective, reducing youth deprivations in well-being may require further investments. Policy makers may not be able to tackle all deprivations simultaneously and understanding the costs may aid in using available funds and political will efficiently. Allied to this, the costs of deprivations in well-being vary strongly, which can point to policy priorities. Being aware of the variation in the costs helps policy makers to set priority areas.
The costing framework
In theory, costs can be divided into i) actual expenses (financial costs) incurred as a direct consequence of deprivation; and ii) foregone economic possibilities due to underutilisation of resources (opportunity costs). In practice, the financial costs are not systematically measured. Further, costs can be attributed to the paying party: the individual experiencing the deprivation and society as a whole, including individuals upon whom negative outcomes are inflicted.
This costing framework focuses on opportunity costs. The financial costs usually lack good-quality data by which to estimate them; either they do not exist or they are not disaggregated by age. Unbalanced data availability can distort cost estimation; well-being deprivations with little data will appear to have lower costs than well-being deprivations with rich data. To avoid these potentials for misrepresentation, this framework focuses on opportunity costs, which can be easily calculated with the available data. Still, this framework does give some examples of financial costs, while recognising that data availability impedes the calculation of most.
This costing framework represents a conservative approximation of the real costs of young people’s well-being deprivations. A robust framework requires that costs can be clearly quantified, monetised and attributed to a well-being deprivation. Numerous intangible costs are associated with well-being deprivations, such as loneliness, psychological distress, lower civic and social engagement, young people opting out of democratic participation, the transmission of poverty and vulnerability to future generations, and emigration. Most of these intangible costs are hard to monetise due to their complexity. They also result from a combination of factors beyond a particular deprivation in well-being. These intertwined causes impede clear cost attribution. Given its difficulty, this exercise restricts itself to calculating foregone labour earnings. Therefore, the estimates have to be seen as a lower bound, whereas the real costs to society will be higher.
Well-being deprivations can have short-term and long-term effects. The costs of short-term effects occur solely in the year of observation. Negative outcomes with long-lasting effects are calculated over a lifetime, discounted for the year of observation and transformed into a year-on-year average over a lifetime. An important underlying assumption to calculate lifetime costs is that the present earnings of any given age group are a proxy for the earnings of a present young person when reaching that age group (World Bank, 2003).
Marginal costs can be more telling for agenda setting. In many cases, eliminating the well-being deprivations of all young people and restoring their full productivity remains an ideal, hard-to-reach scenario. Knowing the costs of well-being deprivation can thus help to set priorities. The costs of deprivations in well-being vary strongly. At times, this variation stems from the number of young people experiencing the different deprivations in well-being. Potentially, this can lead to misinterpretation and underestimating the severity of a well-being deprivation. The marginal costs – the unit cost to each young person experiencing the deprivation – are therefore more important for agenda setting (Eurofound, 2012). Marginal costs can also be used in cost-benefit analyses, e.g. the marginal costs multiplied by the number of reduced deprivations can be contrasted to the programme’s costs.
The subsequent part presents the methodology to estimate the costs of the following well-being deprivations: i) early school leavers; ii) NEET; iii) teenage pregnancy and early motherhood; iv) substance abuse; v) sexually transmitted infections (STIs); and vi) crime and violence. Each is delineated in three parts: i) an explanation of which costs are considered; ii) formulas to estimate costs that can be reliably monetised, quantified and attributed; and iii) suggested potential data sources.
This costing framework uses labour earnings to approximate the costs of well-being deprivation. The underlying assumption of opportunity costs is that the deprived young person would have been working if the deprivation would not have existed. Thus, the deprived individual foregoes earnings due to reduced labour market activity caused by the well-being deprivation. In many developing and emerging economies, self-employment plays a greater role than in OECD countries. Therefore, this framework takes an inclusive approach to income, including income from wages and salaries, as well as income from self-employment and other relevant sources.
This framework is a static model, allowing for an easy application. It disregards spill-over effects of well-being deprivations. More importantly, it does not include labour market responses to the well-being deprivations. An influx of additional young people into the labour force (those currently excluded due to their deprivations) will affect the equilibrium wage. This changing equilibrium wage is, however, not accounted for in this costing framework, which serves as a crude proxy.
The non-zero earnings of young people not affected by the particular well-being deprivation serve as an estimate for the costs of deprivation. The best estimate of the foregone earnings is obtained by applying the propensity score matching method (described in Module 4). This method compares the young people experiencing the well-being deprivation with their statistical twins who are not experiencing the deprivation. Using the earnings of the statistical twin has the advantage that it accounts for potential systematic differences in the background characteristics of the two groups (e.g. ethnicity, skills, educational background, gender and health status), which may bias the potential earnings of the young people experiencing the deprivation. In the absence of adequate data to conduct propensity score matching, wage modelling (e.g. a Mincer-type earnings regression) or non-zero median youth earning can be used to estimate the foregone earnings. This estimate for the foregone earnings should, however, be seen as an upper band, as young people experiencing the deprivation may differ from the non-affected youth in a number of ways and therefore be unable to obtain their median earning (The Prince’s Trust, 2007).
The costs of early school dropout
School attendance is of great consequence to the individual and to society. At school, young people gain knowledge and learn how to function as a member of society. Endowing young people with the most basic skills significantly improves chances of employment in adulthood. Dropping out of school early, combined with other risk factors, can have a domino effect, leading, for example, to substance abuse. It also increases the likelihood of unemployment, which in turn increases the state’s social spending (Assunção and Carvalho, 2003). Furthermore, early school leavers are less likely to find legitimate work, leaving the door open to delinquencies (Lochner and Moretti, 2004). This domino effect underlines the importance of investing in youth to prevent leaving school early.
This costing framework focuses on the negative outcomes of dropping out of school early. Early school dropout is itself a negative outcome and simultaneously a risk factor leading to risk-taking behaviours (see Module 1). Calculating the costs of early school dropout faces the difficulties in quantifying and monetising potential secondary costs. The present methodology leads to a conservative calculation, representing a lower bound of the costs. The costs of risk-taking behaviours due to dropping out of school early (e.g. drug abuse) or subsequent costs (e.g. unemployment) are calculated in the respective sections, if the negative outcomes materialise.
Generally speaking, interrupted human capital accumulations and the lower productivity and income associated with it are the costs of leaving school early (Assunção and Carvalho, 2003) (Table 7.1). Lower productivity compared to peers who complete their education corresponds to lower salaries and fringe benefits (e.g. subsidised meals, company discounts), and consequently to lower taxes revenues.
The total cost of dropping out of school early is the sum of the different cost elements presented below. The costs are calculated over a lifetime. These are divided over the working years to obtain the annual costs over a lifetime. The individual opportunity costs of dropping out of school early are calculated over a lifetime because early school leavers rarely finish school later in their lives “affecting their entire earnings path for the rest of their lives” (Chaaban and Cunningham, 2011).
The individual opportunity costs of leaving school early – the lower productivity of drop-outs – are proxied by the foregone earning potential () and the foregone fringe benefits ( The foregone earning potential is measured by the lifelong differences in earnings between the achieved level of education (es) and the next higher level of education (es’). The earnings of the next higher educational level (es’) are adjusted for an ability bias. It is commonly argued that school drop-outs “are less able” than those who actually completed a higher level of education (Chaaban and Cunningham, 2011). The es’ are therefore commonly reduced by 10% to account for this ability bias. The comparisons undertaken are lower secondary school degree versus primary school degree and primary school degree versus no education at all. The total lifetime earnings are calculated from the drop-out age (t) until the average retirement age (T) and discounted at the rate r. For simplicity reasons, zero years of unemployment are assumed. The presumed age of drop-outs entering the labour market is the average age of graduates of the next higher degree plus two years (Assunção and Carvalho, 2003). Finally, to obtain the total, the costs of the above-mentioned comparisons (e.g. lower secondary degree versus primary school degree) are multiplied by the number of early school leavers at the respective level (ds):
The second element of the individual opportunity costs are the foregone fringe benefits (, which tend to increase with educational level. The foregone fringe benefits are the difference between the average fringe benefits at the actual educational level (fbs) and the average fringe benefits at the next higher educational level (fbs’). The foregone fringe benefits are, like the foregone earnings, lifetime costs and thus have to be discounted over a lifetime. The multiplication by the number of early school leavers at each educational level (ds) gives the total costs:
The society’s opportunity costs are proxied by lower tax revenues due to lower earnings (). The foregone tax revenues assume both full compliance with the tax code and zero tax evasion. The calculation of the foregone tax revenues takes the formula for the foregone earning potential and multiplies it by the marginal tax rate and social security contributions (tx),
The costs of NEET
Among other things, the rate of the young people not in education, employment or training (NEET) depicts the (potential) underutilisation of youth labour. The NEET rate can be more accurate than the unemployment rate because it also includes all young people not actively looking for employment and it excludes those attending school or (vocational) training (as opposed to the inactivity rate). The NEET rate also helps to capture gender disparities, such as different labour market participation rates between young men and women. Being NEET is both a negative outcome and a risk factor and can lead to costly coping mechanisms. Being unemployed or a NEET is also one of the major sources of unhappiness (Layard, 2005), which can increase risk-taking behaviours, such as substance abuse, with subsequent costs to individuals and to society.
In addition to the opportunity costs, there are financial costs, such as looking for a job (going to job centres, screening vacancies, travelling to interviews), and providing (re-) training programmes and active labour market policies targeted at young people. These financial costs are, however, difficult to measure.
The costs of the NEET are calculated only for the year of observation. Although young people enter and exit the labour market frequently, calculating the deterioration of skills is a complex undertaking. Thus, it is assumed that skills deterioration does not have lifetime consequences on labour income (Chaaban and Cunningham, 2011). Evidence of young people’s deterioration of skills and a lasting wage penalty caused by their frequent entry and exit in the labour market are mixed. The lasting impact depends greatly on the individual’s skill level, the duration of being in neither education, employment nor training, and the timeframe of analysis (The Prince’s Trust, 2007).
The costs of the well-being deprivation are calculated for all young people who were in neither employment, education nor training for 6 months or more in the past 12 months (Eurofound, 2012). This threshold is chosen as often there is a natural brief gap between the transitions from education to labour or between jobs. Young people who are without work for less than six months should not be considered NEET, to avoid an overestimation of the costs. The total cost of NEET is the sum of the different cost elements presented below.
The individual opportunity costs of being a NEET (IOCNEET) are the foregone earnings (ey) and fringe benefits due to inactivity. These two elements are multiplied by the number of NEET (NEET) to obtain the total cost. The fringe benefits (fby) granted to young people include all non-wage labour market remuneration.
The society’s opportunity costs of joblessness (SOCNEET) are the foregone average tax payments and the foregone social security contribution (scy), which are obtained by multiplying the annual youth earnings (ey) by the marginal tax rate and social security contributions (tx), and by the youth multiplied by the number of NEET:
The total cost of NEET in Viet Nam is estimated at 2.5% of GDP. The individual opportunity cost (IOC) of NEET compares the earnings and fringe benefits a youth would get if he/she were not a NEET assuming they would miss out entirely on these benefits. For Individual financial cost (IFC), the same is done with out of pocket payments. Society’s opportunity cost (SOC) is calculated by taking youth tax payments of non-NEETs and assumes that this is what society is not getting from NEETs. The society’s financial cost (SFC) is ideally meant to capture the expenditure on training and unemployment benefits for NEETs, although this cost is not available due to data limitations.
The individual opportunity cost due to foregone earnings is again the most significant cost. Not being in employment, education or training means these young people are missing out completely on earnings whilst being NEET, hence are foregoing a lot of money. While the individual finance cost and social opportunity cost is relatively low in comparison (although both account to over 100 million USD each). The individual finance cost is made up of medical expenses. Considering young people incur less medical costs than people older than themselves (Alemayehu and Warner 2004), this could explain why the value is relatively low. The social opportunity cost, which takes into account taxes, is low as some of the equivalent young people being compared to these NEET are those who are in education and training. Thus, they will not be paying much tax.
The costs of teenage pregnancy and motherhood
Teenage pregnancy and motherhood can have negative implications for the well-being of mother and child. Pregnancy during adolescence carries a number of health risks for both. Abortions can pose psychological costs to the teenager, and there are costs of the unborn life. Teenage mothers can be socially stigmatised, and if the child is born out of wedlock, the young mother can experience fewer marriage possibilities. These deprivations in well-being can ultimately result in a disadvantaged upbringing for the child, which will impact the inclusion of the child in society and labour market. Teenage pregnancy and motherhood also reduces inter-generational mobility for the teenage mother and the child (Smeeding, 2015). Teenage mothers tend to suffer constrains that make it difficult for them to get ahead economically; consequently, they also invest less in their children’s opportunities. Although these intangibles (psychological costs, disadvantaged upbringing of children and inter-generational mobility) are part of the costs of teenage pregnancy, monetising and quantifying them is beyond the scope of this exercise.
Teenage pregnancy and motherhood tend to have long-term impacts on the mother’s earnings. Scholars commonly “assume a constant wage gap over their working lives between young mothers and young women who postponed their childbearing” (Chaaban and Cunningham, 2011). This wage gap cannot be solely explained by dropping out of school. Young mothers also have fewer employment opportunities due to their care tasks – especially if they are single mothers – forcing them to take less lucrative jobs, work fewer hours and consequently accumulate less from on-the-job training (World Bank, 2003). Some of the negative outcomes of teenage pregnancy and motherhood, such as dropping out of school or leaving the labour market, may have already materialised. In cases the pregnant teenager or teenage mothers is a NEET, the costs due to foregone earnings are included in that section.
The financial cost of teenage pregnancy and motherhood are the most evident costs, although difficult to measure. Financial costs, such as medical expenses for teenage pregnancies, those mothers and their children, as well as for abortion, tend to be higher than for adult pregnancies due to the higher probability of complications (Assunção and Carvalho, 2003).
The total cost of teenage pregnancy and motherhood is the sum of the different cost elements presented below. As indicated, young mothers face lifelong opportunity costs. These are divided over the working life to obtain the annual costs over a lifetime. This section focuses on young women aged 15-19.
The lower employability of young mothers represents the individual opportunity costs of teenage pregnancy and motherhood ( and are proxied by the earnings gap between young mothers and women who delayed pregnancy. This earnings gap is calculated the same way as the foregone earnings due to early school leaving. It is the difference between the median earnings of women delaying childbirth () and the median earnings of women bearing children during adolescence (). This difference is multiplied by the number of teenage mothers and pregnant teenagers (my).
The society’s opportunity costs of teenage pregnancy and motherhood () are captured by the lower tax revenues due to young mothers’ earnings gap. The calculation follows the same formula as the SOC of early school leaving.
The costs of illicit substance abuse
Illicit substance abuse (marijuana, amphetamines or methamphetamines, cocaine, solvents or inhalants, ecstasy, heroin) by youth has short-term and long-term costs. The long-term costs are the “deterioration of health status as well [the] derangement of mental faculties” (Assunção and Carvalho, 2003). The deterioration of health can also develop long after initiating substance abuse. Although these late costs stem from a risky behaviour started as a young person, they are excluded from this exercise. This costing framework focuses on the opportunity costs resulting from reduced productivity or premature death due to substance abuse.
Illicit substance abuse also has short-term financial costs originating from coping mechanisms, among others. Substance abusers may have health problems or go to rehabilitation centres. The financial costs from medical care and rehabilitation are partially borne by the substance abusers (or the addict’s kin) and partially by society. Substance abusers are also more likely to depend on social assistance. Costs of substance abuse may also impact third parties. In some cases, substance abuse can lead to accidents or violent acts resulting in injury to others or property damage, as well as incarceration (these costs are an outcome of crime or violence and are therefore included in that calculation). The fight against drugs is also a financial cost of substance abuse-related crime and violence. Other third party costs are the psychological distress relatives may suffer, as well as the neglect or psychologically mistreatment substance abusers may inflict on their children. Children of young substance abusers, who grow up in a poor and abusive environment, may face difficulties in integrating into society and the labour market.
The cost of substance abuse is the sum of the different cost elements presented below. Some of the costs are calculated over a working life. These are divided over the working life to obtain the annual costs over a lifetime. The annual costs over a lifetime and the costs only occurring in the year of observation can then be aggregated for illustrative purposes to obtain the cost of substance abuse in the year of observation.
Substance abusers forego earnings while in rehabilitation centres, representing one of the individual opportunity costs (). The foregone earnings due to being in a rehabilitation centre are calculated for the year of observation by multiplying the median youth earnings (ey) with the number of young people in rehabilitation centres (rpy).
Individual opportunity costs can also occur over a lifetime, caused by premature death or reduced productivity (). The foregone earnings of premature death due to substance abuse are the discounted average lifetime earnings. This proxy is a conservative estimation of the costs of premature death. The average lifetime earnings (et) are multiplied by the number of young people dying from substance abuse (sdy). Foregone lifetime earnings potential due to lower productivity are calculated the same way as the foregone earnings due to school drop-out. It is the difference between the median earnings (et) and the median earnings of people with lower productivity due to substance abuse (), multiplied by the number of young substance abusers with reduced physical capacity due to substance abuse (ay). In case of lacking data on the median earnings of people with lower productivity due to substance abuse, the median earnings (et) can be multiplied with a factor accounting for the reduced productivity. Assunção and Carvalho (2003) propose a factor of 9.9%. For simplicity, it is assumed that t is age 30 and that people work until retirement age.
The society’s opportunity costs are the foregone tax income due to rehabilitation centre stays, lower productivity and death. The foregone taxes due to rehabilitation centre stays () are the tax payments of young people (ey* tx) multiplied by the number of young people in rehabilitation centres (rpy).
The foregone taxes due to premature death are the discounted lifetime tax payments (ey * tx) multiplied by the number of young people dying from substance abuse (sdy). The reduced tax revenues due to lower productivity are the difference between the average (tpt) and the reduced tax payments () over a lifetime, multiplied by the number of young substance abuser (ay) with lower productivity due to substance abuse.
The costs of sexually transmitted infections (STIs)
Early sexual activity and unprotected intercourse – whether due to lack of information, lack of access or carelessness – can result in sexually transmitted infections. STIs – especially fatal STIs, such as HIV/AIDS – represent a cost to the infected individual and to society. Some costs, such as reduced life quality or premature death, require the difficult task of monetising the value of life. This framework limits itself to foregone labour earnings and tax revenues due to death and reduced productivity.
Financial costs of STIs are typically the additional costs for medical care. Fatal STI have also indirect costs, such as social stigma, orphaned children or the family distress in case of death. Individuals suffering from STIs contribute less intangible goods (i.e. culturally, politically, etc.) to civil society, and there is the immanent risk for infecting others. Although psychological distress and infection of others are part of STI costs, monetising and quantifying them is beyond the scope of this exercise.
The total cost of STI is the sum of the different cost elements presented below. The costs are calculated over a working life. These are divided over the working life to obtain the annual costs.
Young people suffering from STIs forego lifetime earnings due to lower productivity and premature death. These individual opportunity costs ()are calculated the same way as the foregone earnings due to substance abuse. It is the difference between the median earnings (et) and the median earnings of people with lower productivity due to STIs (), multiplied by the number of young people with reduced physical capacity due to STIs (). For simplicity, it is assumed that t is age 30 and that people work until retirement age. Approximating the costs of premature death due to fatal STIs as the discounted lifetime earnings is a conservative approach, which underestimates the true cost. The average lifetime earnings (et) are multiplied by the number of young people dying from STIs (si).
The society’s opportunity costs are the foregone taxes () due to dying prematurely and lower tax revenues due to reduced productivity. The foregone taxes due to premature death are the discounted lifetime tax payments () multiplied by the number of young people dying from STIs (siy). The reduced tax revenues due to lower productivity are the difference between the average () and reduced tax payments () over a lifetime, multiplied by the number of young people with STIs with lower productivity (dy).
The costs of crime and violence
Careers in crime usually lead to further exclusion. Young offenders, especially if incarcerated, tend to experience educational underachievement and deterioration of human capital, unemployment, as well as exclusion from participation in society (The Prince’s Trust, 2007). Crime and violence therefore constitutes a serious impediment to young people’s well-being. Although it is difficult to determine whether the deprivations stem from engaging in criminal activities or whether young criminals have less favourable characteristics to begin with, it is clear that incarceration amplifies young people’s deprivations.
Incarceration or lower productivity due to crime and violence represent some of the opportunity costs. Other opportunity costs are the crime victims’ foregone earnings due to premature death or lower productivity resulting from lasting injuries of an assault. Foregone earnings consequently result in foregone tax revenues for the state. Given the youth focus of this costing framework, the victim or perpetrator have to be young.
In some countries, juvenile delinquency and violent gang activity pose serious problems to society, creating high financial costs. Victims of crime have to restore physical damage to their property and, in some cases, cope with psychological damage (e.g. enter therapy). Society assumes the costs of incarceration, as well as the legal and medical expenses due to damages or injuries to others and/or state property. High rates of crime and violence – or the perception of high rates – push victims and those afraid of becoming victims to spend on private security measures (e.g. security cameras, guards, safes, etc.). These private investments in security measures are in addition to the costs of public security. A heightened sense of insecurity also has more far-reaching consequences. Productivity can decrease at a macro level, and revenue generators, such as tourism, can drop off. These potential foregone earnings are disregarded in this exercise as their causes are difficult to attribute and the costs are difficult to quantify.
The total cost of crime and violence is the sum of the different cost elements presented below. Some of the costs are calculated over a working life. These are divided over the working life to obtain the annual costs over a lifetime. The annual costs over a lifetime and the costs only occurring in the year of observation can then be aggregated for illustrative purposes to obtain the total cost of crime and violence in the year of observation.
The perpetrators’ foregone earnings due to incarceration are one of the individual opportunity costs (. The costs due to incarceration are the median youth earnings (ey) multiplied by the number of young people in prisons (py). For simplicity, it is assumed that being incarcerated has no lasting impact on lifetime earnings.
The foregone lifetime earnings due to premature death or reduced productivity are the other individual opportunity costs (. The costs of premature death of perpetrators are approximated by the median lifetime earnings (et) multiplied by the number of young perpetrators dying (). Perpetrators’ foregone lifetime earnings due to lower productivity are calculated the same way as the foregone earnings due to substance abuse. They are the difference between the median earnings (et) and the median earnings of people with lower productivity due to reduced physical capacity due to their criminal actions (), multiplied by the number of young perpetrators with reduced physical capacity due to their criminal actions (dy). For simplicity, it is assumed that t is age 30 and that people work until retirement age.
Victims of youth crime also forego earnings due to premature death or lower productivity; these costs are part of the society’s opportunity costs (. The foregone earnings of casualties due to juvenile crime are the discounted average lifetime earnings (et), multiplied by the homicide rate due to juvenile crime (). The formula of the foregone lifetime earnings due to lower productivity of the victims is identical to the perpetrators’ formula but replacing dy with the number of victims of juvenile crime with reduced productivity (v).
Society forgoes tax revenues due to incarceration, premature death and reduced productivity. The foregone taxes of the youth prison population are the median tax payments of young people (tpy) multiplied by the number of incarcerated young people (py).
The foregone taxes from homicides are the lifetime tax payments multiplied by the homicides ( and ). The reduced tax revenues due to lower productivity follow the same logic as the reduced tax revenues due to substance abuse — multiplying that difference by the number of perpetrators (dy) and victims (v) with lower productivity.
References
Assunção, J. and L. Carvalho (2003), “Brazilian Youth at Risk: Estimating the Costs of Not Preventing Risky Behavior”, Background Paper (unpublished), World Bank, Washington, DC.
Boarini, R., A. Kolev and A. McGregor (2014), “Measuring Well-being and Progress in Countries at Different Stages of Development: Towards a More Universal Conceptual Framework”, OECD Development Centre Working Papers, No. 325, OECD Publishing, Paris, https://doi.org/10.1787/5jxss4hv2d8n-en.
Chabaan, J. (2008), The cost of youth exclusion in the Middle East, The Middle East Youth Initiative working paper, Washington DC
Chabaan, J. and W. Cunningham (2011), Measuring the Economic Gain of Investing in Girls: The Girl Effect Dividend, World Bank, Washington, DC.
Cohen, M.A. (1998), “The Monetary Value of Saving a High-Risk Youth”, Journal of Quantitative Criminology, Vol. 15/5, Springer Publishing, New York, pp. 5-33.
Eurofound (2012), NEETs – Young people not in employment, education or training: Characteristics, costs and policy responses in Europe, Publications Office of the European Union, Luxembourg
Jütting, J. and J.R. de Laiglesia, (eds.) (2009), Is informal Normal? Towards More and Better Jobs in Developing Countries, OECD Development Centre, OECD Publishing, Paris.
Layard, R. (2005), “Happiness”, The Penguin Press, New York
Lochner, L. and E. Moretti (2004) “The Effect of Education on Criminal Activity: Evidence From
Prison Inmates, Arrests and Self-Reports”, American Economic Review, 94, 155-89.
OECD (2014), Social Cohesion Policy Review of Viet Nam, OECD Publishing, Paris, https://doi.org/10.1787/9789264196155-en.
OECD (2011), How’s Life?: Measuring Well-being, OECD Publishing, Paris, https://doi.org/10.1787/9789264121164-en
Prince’s Trust (2007), The cost of exclusion: Counting the cost of youth disadvantage in the UK, Prince’s Trust, London.
Smeeding, T. (2015) “Inequality and intergenerational mobility: Looking ahead, not behind” OECD, Presentation at New Approaches to Economic Challenges Seminar. Paris, France. 13 April 2015. World Bank (2008), Youth at Risk in Latin America and the Caribbean: Understanding the Causes, Realizing the Potential, Directions in Development: Human Development, World Bank, Washington, DC.
World Bank (2004), Young People in South Eastern Europe: From Risk to Empowerment, World Bank, Washington, DC.
World Bank (2003), Caribbean Youth Development: Issues and Policy Directions, World Bank, Washington, DC.