The Samartian’s Dilemma & Welfare

selective focus photography cement
Photo by Rodolfo Quiru00f3s on


On a philosophical level, the virtue of charity is highly exalted. Few are willing to question if charitable efforts are effective. An even increasingly small number of philanthropists are willing to entertain the notion that they are harming the recipient. It is counter-intuitive to fathom the idea of our helping hand is causing negative consequences. Unfortunately, sometimes good intentions do not yield good results.

Despite the clear need for charity in our society that still doesn’t negate potential externalities. Intentions are mutually exclusive from the repercussions. This observation is demonstrated by Noble Laureate James M. Buchanan’s Samaritan’s Dilemma.  Buchanan based his postulation upon the Parable of the Good Samaritan. This biblical passage details an account of  “… the Samaritan aiding a man who had been beaten and robbed by thieves…” ( Wagner, P.11) [1]. Buchanan took this biblical allusion one step further and generalized it to charity as a whole. Once the recipient is cognizant that the charitable donation is larger than what they can earn from labor, they will lose the incentive to work (Wagner, P.11) [1]. Thus, they become dependent upon charity.

While this concept is applicable to private charity, it is quite prevalent in public policy. Many “safety-net” programs are merely forms of socialized charity. The implication of “socialized charity” is intended from a value-neutral standpoint. The Samaritan’s Dilemma is not confined to our stereotypical examples of government assistance. For instance, the Samaritan’s Dilemma arises in the area of subsidies for crop insurance. A 2009 study found a” … 0.2% decrease for every percent increase in expected disaster payments…” (Deryugina & Kirwan, P. 27) [2].  Similar findings are also applicable to moral hazard in flood insurance policies.  As policymakers succumb to the pressure of voters demanding increased funding for flooding relief. The number of property owners with flood insurances decreases (Michel-Kerjan et al, P. 9) [3].

The presence of moral hazard seems to be prevalent throughout disaster relief literature. Applying both domestically and internationally. If the precedent for government aid is established individuals are less apt to take precautions. Making them solely dependent on aid provided by various government agencies.

Why Focus On Means-Tested Welfare Benefits? 

It is quite evident that there is a robust body of literature related to disaster relief and the Samaritan’s Dilemma. The obvious question becomes, why focus on means-tested welfare?  For one, programs such as SNAP, WICC, Medicaid, etc. tend to find into our stereotypical perception of charity. Making it a clear example of where the line of assistance veers into incentivizing long-term dependence. Also, funding for such programs is always a consistent talking point pertaining to economic policy. There is also a lot of misconceptions concerning the incentives for welfare dependence. The ubiquitous image of indolent dirtbags avoiding work is largely a myth. The vast majority of welfare recipients wish to participate in the labor force. However, typically the total value of welfare benefits allocated tends to surpass an annual salary for most entry-level positions.  A fact that has been demonstrated by a 1995 study (Tanner et al. p.2) and the 2013 follow-up (Tanner & Hughes, p. 7) conducted by the Cato Institute [4] [5].

The Cato Institute 1995 Welfare Study:

The 1995 welfare study conducted by the Cato Institute found that overall combined welfare benefits generally exceed the income of most entry-level positions. The added benefit being even a state with benefits comparable to minimum-wage such allocations would be tax-exempt (Tanner et al. p.1) [6]. Clearly aligning incentives with long-term welfare dependence.  In the short-term could be perceived as being within the best interest of a rational actor. Setting the stage for applicability of the Samaritan’s Dilemma in the sphere of entitlement programs.

In order to fully illustrate how welfare benefits reduce incentives to work, it is best to compare wages with the sum of allocated benefits. Back in 1995, 40 states’ welfare benefits collectively amounted to more than $8.00 per hour (Tanner et al. p.1) [7]. In 17 states, the total amount was equivalent to more than $10.00 per hour (Tanner et al. p.1) [8]. Some of the more generous states in the country (HI, AK, MA, CT, NY, NJ, RI,& Washington D.C.) paid out more than the equivalent of $12.00 per hour in welfare benefits. That is more than 2.5 times the federal minimum wage in 1995 (Tanner et al. p.1) [9]. 8 states within this time frame provide more than $25,000.00 in welfare benefits annually (Tanner et al. P.2) [10].  This is equal to the purchasing power of $42,154.53 in 2020 when adjusted for inflation.

It is imperative that we cross-compare these findings with average entry-level wages back in 1995. This will establish a proper context for the previously detailed allocations.  Simply stating that Hawaii paid out the wage equivalent of $17.50 means very little without the appropriate context of market wages (Tanner et al. P.2) [11]. Per Tanner et al. :

An entry-level secretary can expect to earn about $9.01 per hour, which is less than equivalent to welfare benefits in 29 jurisdictions.[38] The national median wage for a janitor is $6.75 per hour.[39] Welfare recipients in 47 jurisdictions receive more in benefits than the average janitor. It is important to realize that, contrary to popular belief, the average janitor’s wage is well above the minimum wage rate of $4.25 per hour. Perhaps more interesting, the national median wage for computer programmers is about $13.03 per hour–less than the welfare benefit levels in the six most generous states.[40]  (Page 23-24) [12].

It is quite conspicuous why a welfare recipient would be susceptible to indefinite reliance on entitlement programs. If you are able to obtain more than the equivalent of a computer programmer’s median annual salary in public assistance it would be irrational to work. Typically those receiving government assistance lack the human capital to make equivalent wages in the labor market. When you combined the advantages of these benefits being tax-free and the general human preference for leisure, the choice is simple. Continue to receive welfare benefits as long as possible. Not only are incentives to work are removed, but consequentially the drive to improve one’s circumstances is relinquished. An occupation such as computer programming requires a lot of job training and even college degrees. Keeping human nature in mind, if there is a way to circumvent the schooling and job training but have equal compensation most people will take the shortcut. It is not laziness or malevolence, but rather a rational cognitive appraisal. That is why merely  68.6 % of welfare recipients in 1995 were actively seeking employment (Tanner et al. P.2) [13]. Poor people aren’t stupid, they respond to incentives in the same way anyone else would.

It should be important to not overgeneralize the findings of the 1995 study. The potential for misapplication is highly probably. Both by welfare advocates as well as opponents. The study notes how there were 77 programs in 1995, no single recipient receives benefits from each one (Tanner et al. P.4) [14].  For example, only 23 % of  AFDC beneficiaries had received housing assistance (Tanner et al. P.26) [15]. However, most had received benefits from multiple programs.

The 1995 study concludes that the salient incentives for long-term welfare dependence are tied to the number of benefits provided. As a means of policy reform, it is recommended that welfare benefits be reduced to minimize persistent reliance on such programs (Tanner et al. P.30) [16]. The timing of this study is somewhat ironic considering the timing of the study in relation to the 1996 Welfare reform of the Clinton administration. I would strongly surmise that this study provided some fodder for justifying such a policy change.

The Cato Institute 2013 Welfare Study:

In social sciences much like other disciplines that utilize empirical research replication of results is paramount.  In science, if results cannot be replicated odds are the researcher’s results were due to sampling error. Verifying the importance of follow-up studies. Michael Tanner and Charles Huges conducted a follow-up to the 1995 study.  One core difference between the 1995 and 2013 study is the implementation of the Welfare Reform Act of 1996. It would be natural to assume that the parameters of this reform have reduced the numbers of people dependent on welfare. Unfortunately, such an assumption is superficial and inaccurate.

Despite the efforts of the Clinton administration, the welfare reform of 1996 wasn’t as successful as anticipated. The dollar value of welfare benefits paid out has declined in 18 states since the reform (Tanner & Hughes, P. 8) [17]. However, in 33 states and the District of Columbia, the monetary sum of benefits paid out increased even when adjusted for inflation (Tanner & Hughes, P. 8) [18]. Placing into question the overall impact of the 1996 reform. Since 1995 the number of federal programs geared towards low-income recipients ballooned 126 (Tanner & Hughes, P. 7) [19]. Which does not lend much confidence in the effectiveness of the 1996 reform.

The unfortunate facts are that many state welfare programs are still far too generous. Despite minor improvements.  In 2013, 13 states had benefit allocations averaging over $15.00 per hour (Tanner & Hughes, P. 7) [20].  The study also notes that 35 states still have welfare benefits that exceed the wages earned on a minimum-wage salary (Tanner & Hughes, P. 7) [21]. The subsequent study found that in 12 states leaving welfare would result in a loss of income. Even with EITC and CTC tax credits taken into account (Tanner & Huges. P.7) [22]. An incentive that makes up for taking a job that pays slightly less than welfare benefits in  39 states. However, still cannot counterbalance the value of leisure (Tanner & Huges. P7) [23].

In  2013, the issue of welfare allocations eclipsing many entry-level positions is still a lingering issue.

In 11 states, welfare pays more than the average pre-tax first-year wage for a teacher. In 39 states it pays more than the starting wage for a secretary. And, in the 3 most generous states a person on welfare can take home more money than an entry-level computer programmer. (Tanner & Hughes, P.8) [24].

The number of states providing benefits surpasses the salary of an entry-level computer programmer has been reduced by 50% since 1995. However, it does seem superfluous to have any aid to top the salary of a computer programmer. Computer programming is a prime example of skilled labor. That alone is enough to discourage those in need to find work. Why go to college to gain such skills when you can just apply for welfare? However, it seems as if such generous distributions are relegated to a minority of states. State welfare benefits exceeded the salary of a starting secretary in 29 states in 1995. That number jumped to 39 in 2013. This is obviously a step in the wrong direction. Especially considering secretary positions can be a solid career stepping stone.

The five most generous states all increased the overall welfare allocations since 1995. Hawaii had a  $7,265.00 increase in average welfare benefits paid to a household. Washington D.C. a notably $8,730.00 increase. Massachusetts exhibited a $5,169.00. Connecticut a modest increase of  $1,781.00. Finally, New Jersey with a $5,533.00 (Tanner & Hughes, P. 8) [25]. To truly put these numbers in perspective it would be helpful to see how these benefits rank terms of hourly wages. For Hawaii, it amounts to the pre-tax equivalent of a resounding  $29.13 per hour. Washington D.C. $24.43 per hour. Massachusetts $24.30 per hour. Connecticut $21.33 per hour. New York $21.01 per hour (Tanner & Hughes, P.12) [26].  Please note that in Hawaii minimum-wage as 2020 was $10.10 per hour [27].

While derisive criticism could be directed at the states for generous benefits programs, there are other factors contributing to the lackluster results of the 1996 welfare reform. There is much due opprobrium that can be levied against the work requirements. The 1996 legislation added more stringent work requirements as a contingency for receiving means-tested aid [28].  Per federal law,  each can provide hardship work exemptions for up to 20 % of all welfare recipients (Tanner & Hughes, P. 39) [29].  Such a restriction seems like reasonable restraint many there are still states such as Massachusetts that have few recipients participating in work actives.(Tanner & Hughes, P. 39) [30].  For example, per the study, only 18% of adult participants in TANF were engaged in work actives (Table, P. 41).

Nationally less than 42% of adults receiving public assistance are working (Tanner & Hughes, P. 41) [31]. Even at that, the definition of work activities stretches beyond actual productive labor. Encompassing job training and even schooling. The definition of work activities was made even broader under the Obama administration. Per a 2012 executive order (Tanner & Hughes, P. 41) [32]. Providing more malleability in the parameters for welfare requirements only creates more loopholes. Participants are most likely not looking to cheat the system. However, broadening requirements increases the aptitude and propensity for chronic dependence. In turn, fulfilling the prophecy of Buchanan’s Samaritan’s Dilemma.


A help-handing can make all the difference for those who are in need. I would never set out to write a critical polemic against the virtue of charity. The institution of privatized charity would be the preferential over socialized assistance. Whenever the costs are socialized there is a disconnect between funding and the actual effectiveness of the government program. Most taxpayers operate under the assumption that the policy must be effective. Acquiesce the existence of welfare as a necessity, if it wasn’t the program wouldn’t exist. Few question whether or not their helping-hand is securing the ball-and-chain of multi-generational welfare dependence.

My policy preference concerning welfare programs leans Libertarian. However, the complete abolition of social assistance initiatives is unrealistic. Any politician suggesting such policy prescriptions would be committing political suicide. Most of these programs are entrenched fixtures of the massive bureaucratic body. Making them difficult to dismantle. Odds are most voters’ preferences would align with keeping such efforts in place. Therefore, it is important to look at this issue from the framework of reform versus the lofty aspirations conveyed in political theory.

Once you re-orientate your expectations become a matter of honing in welfare. Focusing on shaping these programs into being temporary assistance versus lifelong entitlements. Scaffolding towards independence versus multi-generational shackles. The only way to make this policy shift would be through reducing the value of welfare benefits paid. In turn, such a policy adjustment would realign motives towards investment in human capital and situational improvement. Not indefinite continuation of the same programs. Which is why I would recommend time limits on the duration of time a recipient can receive benefits. It is evident that work requirements are flimsy barriers. Flimsy barriers which can be easily penetrated.

Judging by some of the gaping holes in the Welfare Reform Act of 1996, it is reasonable to question if the Clinton Administration was serious about welfare reform. Easy to see this piece of legislation as a series of insincere concessions between a Democratic White House and a Republican congress. Even in light of the loopholes in the initial legislation, the reform initiative has been further eroded by subsequent administrations. Per the Cato Institute 2013 study pertaining to the work requirement changes made by the Obama administration:

However, other observers disagree, pointing out that the definition
of work activities is already extremely loose so that any increased latitude can only make the situation worse. Ron Haskins, who as a Republican committee aide helped draft the 1996 welfare reform and who now is an analyst for the Brookings Institution, says that if the administration “wanted to undermine the work requirement,” the new policy “is away to do it.” (Tanner & Hughes, P.45-46)

Which circles back to the whole concept of welfare dependence being a byproduct of legislative flaws. Designing welfare programs that make it too easy or too rewarding to obtain welfare benefits will breed long-term dependence. The question becomes are we are helping or hurting these individuals. I would be speculating we are hurting them. They continue to lack the skills for substantial employment. All it could are budget cuts or elimination of a program and then they are living in a cardboard box. The price of being completely reliant on government assistance is having no autonomy over your economic future. No hope of upward mobility. Being held captive by congress and voter attitudes towards welfare. If either falls out of favor of supporting welfare, vicissitudes do not bode well for recipients.

Milton Friedman’s Answer to Welfare Reform

adult beanie crisis despair
Photo by Pixabay on



It is quite evident that the welfare programs in the United States are beyond broken. So much that many of these bloated and red-tape draped initiatives are seem unsalvageable, irredeemable, and impossible to adequately reform! Not only do they often fail to lift recipients out of penury appears to have generated cycles of multi-generational poverty. Leading to the perception that these programs operate more as institutional forms of job security for bureaucrats rather than measures to help the poor. Forms of public assistance stressing dependence versus progressive graduations of increase autonomy are a hair away from being medieval serfdom. A pen stroke away from the restoration of the repugnant practice of sharecropping. If being shackled to the debts incurred at the company store wasn’t haunting enough the ghosts of past sins don’t stop there. Sharecropping often occurred on former slave plantations in the Southeastern region of the United States. Freed only to remain on the same servile living quarters of your captive ancestors. This analogy is fitting because the current welfare system is analogous to an implicit form of slavery, like serfdom or sharecropping.


When faced with the social and financial costs of the current programs it is extremely tempting to advocate the flat out abolition of all forms of welfare. It is ineffective at helping recipients claim up the socio-economic ladder. Servers more as a temporary supplement in contrast to a permanent solution. The amount of money that has been squandered on such ineffective measures is very concerning. Considering the dismal results and the ever-expanding expenditures allocated for welfare programs, it is fair to surmise that government institutions are inept at efficiently running them. Despite the qualms any fiscally-minded voter may have pertaining to social welfare, outright elimination is unrealistic.  The public outcry would be so profuse and pervasive it would not be a pragmatic political reality. I hate to expound upon such a reaction in such cynical terms; however, one quirk of human nature is that people like to obtain goods and services for free. Even in the private sector, if a product or service is being given away for free; people are willing to kill each other to obtain it. The best you can hope for is a significant downgrade in the amount of bureaucracy and spending utilized to dispense welfare services.

The question becomes if we cannot feasibly relinquish all welfare programs how do we implement a more efficient and effective means of providing such services?  Literally a trillion-dollar question. Nevertheless, a question that we need an answer too, especially in the age of uninhibited profligate spending.  The answer may come from an unlikely source, the Nobel Laurate Libertarian economist Milton Friedman. The answer is the implementation of a Negative Income Tax (NIT). The Negative Income Tax is often confused with Universal Basic Income (UBI), a policy championed by Democratic presidential candidate Andrew Yang. However, UBI is not a means-tested form of welfare entitlements whereas NIT is. In basic principle, NIT works by providing :


“….the percentage difference between an individual’s income and an income cutoff, or the level at which they start paying income tax..” [1].


A firm example of how this would be applied the following example would best demonstrate how this would work.

“The NIT would thus be a mirror image of the regular tax system. Instead of tax liabilities varying positively with income according to a tax rate schedule, benefits would vary inversely with income according to a negative tax rate (or benefit-reduction) schedule. If, for example, the threshold for positive tax liability for a family of four was, say, $10,000, a family with only $8,000 of annual income would, given a negative tax rate of 25 percent, receive a check from the Treasury worth $500 (25 percent of the $2,000 difference between its $8,000 income and the $10,000 threshold). A family with zero income would receive $2,500.” [2].


Even though budget hawks and limited government purists would still scoff at the notion of implementing NIT. Especially individuals who unconditionally view taxation constituting theft. While their apprehensions are understandable, the unfortunate reality is that the likelihood of America ever becoming a “voluntary” society is minuscule. There is too much-invested interest ranging from self-interested bureaucrats to the welfare recipients themselves that would be an impossible sell. It is hard enough to persuade the public to go along with NIT. If implemented, fiscally, this would be a huge victory in reforming welfare.  Per Milton Friedman’s 1962 book Capitalism and Freedom:


“…. In 1961, the government amounted to something like $ 33 billion (federal,

state and local) on direct welfare payments and programs of all kinds: old age

assistance, social security benefit payments, aid to dependent children, general

assistance, farm price support programs, public housing, etc. I have excluded

veterans’ benefits in making this calculation. I have also made no allowance for

the direct and indirect costs of such measures as minimum-wage laws, tariffs,

licensing provisions, and so on, or for the costs of public health activities, state

and local expenditures on hospitals, mental institutions, and the like.


There are approximately 57 million consumer units (unattached individuals

and families) in the United States. The 1961 expenditures of $ 33 billion would

have financed outright cash grants of nearly $ 6,000 per consumer unit to the 10

percent with the lowest incomes. Such grants would have raised their incomes

above the average for all units in the United States. Alternatively, these

expenditures would have financed grants of nearly $ 3,000 per consumer unit to

20 percent with the lowest incomes. Even if one went so far as that one-third

whom New Dealers were fond of calling ill-fed, ill-housed, and ill-clothed,

1961 expenditures would have financed grants of nearly $ 2,000 per consumer

unit, roughly the sum which, after allowing for the change in the level of prices,

was the income which separated the lower one-third in the middle 1930s from

the upper two-thirds. Today, fewer than one-eighth of consumer units have an

income, adjusted for the change in the level of prices, as low as that of the

lowest third in the middle 1930s.


Clearly, these are all far more extravagant programs than can be justified

to “alleviate poverty” even by a rather generous interpretation of that term. A

program which supplemented the incomes of the 20 percent of the consumer

units with the lowest incomes so as to raise them to the lowest income of the

rest would cost less than half of what we are now spending.  (Page 159) [3].”



Clearly, even back in 1961 supplanting the many of the “New Deal-era” entitlements with NIT would substantially save money. Even when adjusted for the rate of inflation, those savings would be even more pronounced in 2019. Within the past 59 years, the size and scope of government have only drastically expanded which is in lock-step with sharp increases in spending. The national deficit back in 1961 was $289 Billion with a Debt/GDP ratio of 52%. [4]. While the years between 1961 and 1988 saw fluctuations, it was not until the mid-1990’s that we started to see profound increases in deficit spending [5]. As of 2018, the Debt/GDP ratio reached 104% [6]. Doubling since referenced seminal work of Friedman has been originally published! If the United State were, in theory, able to reduce all current welfare initiatives it would significantly reduce expenditures on such programs. The excessive amounts of spending on welfare programs is one of my biggest reservations pertaining to such entitlements.


The question becomes is how does UBI stack up in comparison to NIT when it comes to budgetary allocations. As you can imagine UBI is the more expensive of the two policies. What’s the damage?  Per an article from the Niskanen Center: Universal Basic Income is Just a Negative Income Tax with a Leaky Bucket demonstrates this point astoundingly. At a proposed rate of $10,000 per person, annually would result in an estimated expense of $3 trillion dollars. It estimated even if the allocation per person was cut in half it would still constitute most of the federal budget minus “… Social Security, Medicare, defense, and interest payments” [7]. However, it does appear based on a rough estimate of the cost of NIT that the total would be $182 billion annually. Which would be equivalent to the combined cost of SSI, SNAP, and EITC. Obviously, it is still a noteworthy expense but pales in comparison to the cost of UBI implementation and that of our current expenses.  It becomes quite evident how foolish it becomes to utilize the terms UBI and NIT interchangeably considering the gargantuan differences in the costs. The referenced article questions why more policymakers are advocates of UBI over NIT. That is a reasonable question. Especially when one policy is more economically efficient from the standpoint of expenses.


Notable reductions in welfare spending are not the only benefit that would come of narrowing down public assist to only NIT. An article published by the Independent Institute back in 2008: If you Really Want to Help the Poor, Remember Milton Friedman, focuses on these advantages. 1). The article mentions how NIT benefits the individual recipient by receiving more money through not punishing them for working. Many current programs cut benefits once the starts to receive a low threshold of compensation. Relinquishing incentives for working. 2.) It is more efficient than other forms of welfare by eliminating all the departments that currently allocate services. This would amount to a radical consolidation of internal government agencies. 3.) Being the funds are appropriately dispensed. In other words, the money is given to the recipient versus squandered by overhead costs (such as the salaries of administrators) or reallocated for another purpose. It speculated that a mere “… 15 cents of every dollar finds its way to the poor…” [8]. Providing a truly awe-inspiring example of government inefficient at work.


The current welfare system is an abysmal train-wreck that most likely needs to be gutted down to the studs. Completely eliminating it is highly improbable, so the unfortunate fact is society needs to find an efficient compromise. From the standpoint of economic and institutional efficiency, a Negative Income Tax would be the most advantageous option. Not to be confused with Universal Basic Income, similar premise, however, financing that variation of distributed funds is a different story. One reading this article today, please stop substituting UBI for NIT and vice versa. Both policies have different ramifications when it comes to funding. In all honesty, one of the policies seems more rational and effective than the other. While the other is merely a Keynesian stimulus package camouflaged as a theoretical solution from the Chicago School of Economics. NIT only provides subsidies to those of the lower-income strata. UBI, in contrast, provides a siphon to everyone regardless of finical need, which is irrational. It harkens back to the fixation of the Keynesian school when it comes to economic consumption being the core impetus for economic health. Which in all honesty hasn’t always proven to be true. If you are going to redistribute tax dollars you might as well give to those in need rather than haphazardly provide it to everyone on the feeble hope it will spur an uptick in spending. In all honesty, welfare programs were intended to operate as a socialized charity. While I personally believe all charity should be privatized, I understand that I am in the minority holding such a view. If my tax dollars are going to be retributed at the very least do so in a rational manner. A five-time lottery winner doesn’t need any of my tax contributions.


I always try to give people the benefit of the doubt rather than casting suspicion upon them of duplicity. I am perplexed by the fact that Andrew Yang claiming that Milton Friedman was a proponent of UBI. As demonstrated previously UBI and NIT are similar but not the same.  Yang is either ignorant of the difference because he never deeply examined the distinction between the two or he’s dishonest. Either could be the case as he is a presidential candidate. However, the potential for political opportunism is always lurking in the background for anyone with presidential aspirations. I wouldn’t put it over Yang to sell the benefits by omitting the key differences between UBI and NIT and conflating the two intentionally. Then utilizing the connection to Friedman and his premarket legacy to sell the American public on NIT’s more expensive brother. Why sell UBI over NIT? It is a matter of obtaining votes. What is a better sales pitch only low-income Americans get a check or everyone gets a check? The answer to that rhetorical question should be evident. The problem is when people are so engulfed in their own self-interest they are blind to the big picture, they are incentivized to make idiotic decisions.  Long run everyone getting a check is more harmful than regulating it to the poor.