AN EVALUATION OF FINANCIAL INCENTIVE POLICIES FOR ORGAN DONATIONS IN THE UNITED STATES
Abstract
This paper examines the association between financial incentives and organ donations. Although the National Organ Transplant Act of 1984 prohibits financial compensation for organs for transplant, we focus on the impact of laws that influence the relative cost of deceased and live organ donations on the supply of organs for transplant. First, we hypothesize that states that have relatively stringent funeral regulations, which have been associated with higher whole-body donations, will have fewer organ donations. Second, we examine the impact of two common state laws that offer financial compensation to live donors: one that allows a tax deduction for costs incurred and the other which entitles government employees up to 30 days of paid leave. We find no evidence to support that these laws affect organ donations. (JEL I11, I18)
I. INTRODUCTION
It is well documented that the demand for organ transplants far exceeds the supply (see http://www.optn.org). In the 2009 winter, approximately 100,000 people were on waiting lists for organ transplants. The majority of people (over 78,000) were waiting for a kidney transplant. Relying on altruism to meet the need for transplantable organs, as the United States does, has been unsuccessful in meeting the demand. Consequently, a number of different approaches to increase the supply of organs have been proposed. These proposals include: presumed consent laws, as are common in many European countries; first-person consent laws that would end the practice of asking permission of the next-of-kin after the individual has chosen to donate; increased funds for education; implementation of “best practices” in terms of approaching the family of potential donors; increased use of “marginal” donors; and some form of financial compensation.1
Several economists have examined the potential for a market for organs and estimated the implied equilibrium price. Two of these studies (Kaserman and Barnett, 2002;2Wellington and Whitmire, 2007) only consider a market for cadaveric kidneys, and estimate an equilibrium price using survey data. But these studies find huge differences in the estimated equilibrium price. Kaserman and Burnett estimate it to be under $1,000, whereas Wellington and Whitmire estimate the price to be in the billions! Wellington and Whitmire point out that using survey data for this type of study is problematic, and argue that one really needs experimental data. Becker and Elias (2007) construct a theoretical model to estimate the equilibrium price of a kidney, and specifically include live donors in the market. In fact, the authors argue that the equilibrium price will be determined by the supply of live donors. They estimate that an equilibrium price of approximately $15,000 per kidney would compensate live donors for the risk of death, foregone wages, and reduced quality of life.
However, the National Organ Transplant Act (NOTA) of 1984 prohibits financial compensation for organ transplants, and that prohibition is reiterated in the Uniform Anatomical Gift Act of 1987. There is much disagreement by experts (including those from the medical community, ethicists, and economists) on whether financial compensation or a market for organs is “ethical.”3 The chairman of the National Kidney Foundation, Charles Fruit (2006), argued that financial incentives “could move people to view organs as commodities, diminish altruistic donations, and lead to the potential exploitation of lower income individuals.” In contrast, Roth (2007) notes, “In an informal poll following a debate on the subject at a recent meeting of the American Society of Transplant Surgeons, a majority of those polled expressed a willingness to contemplate a trial or demonstration project involving compensation for organ donors (personal communication, Arthur Matas, 1/27/07).” Additionally, the AMA recently urged Congress to amend the NOTA of 1984 to allow for pilot studies with financial compensation for families of cadaveric donors as a means to reduce the shortage of organs available for transplant (PR Newswire, 2008).
Although financially compensating either family members of cadaveric donors or live donors for organs is illegal, there are ways that financial incentives have been introduced, which could directly or indirectly affect the donation of organs for transplant. Harrington and Sayre (2007) examine the impact of funeral regulations on whole-body donations to medical schools, noting that more stringent funeral regulations, and hence more expensive funeral costs, have led to more whole-body donations in order for family members to avoid the higher funeral costs. In terms of organ donation for transplants, the authors point out that since donating organs for transplant typically precludes donating bodies to medical schools, more stringent funeral regulations could also lead to a lower rate of organ donation for transplants.4
In terms of live donations, recent state legislation has allowed donors to be compensated, to some degree, for costs incurred due to an organ donation. These laws are of two general types. (1) It allows individuals to take up to a $10,000 tax deduction for costs incurred due to the organ donation. (2) It targets government employees, and typically allows them up to 30 days of paid leave due to the donation.
This paper examines how state-level funeral regulations and laws compensating live donors, both of which create financial incentives, may affect organ donations for transplants. Are stringent funeral regulations unintentionally reducing the supply of cadaveric transplantable organs? Are states that offer some type of financial compensation for live donors successfully encouraging more live donations? Or are these incentives too small, or simply not relevant, to impact the supply of transplantable organs?
Section II focuses on the impact of differences in funeral laws on cadaveric organ donation, while Section III examines the impact of state laws concerning live donors and their impact on live donations. Section IV summarizes and considers the lessons learned, and the implications for organ transplants in the United States.
II. FUNERAL LAWS AND CADAVERIC DONATIONS
In this section we examine the effect that variations in the opportunity costs involved in organ donation have on the cadaveric donation rate. We use variation in state funeral regulations as a measure of cost of cadaveric organ donations. The logic is that state funeral regulations increase the cost of burial in certain states. In these states, therefore, there is an incentive to donate whole bodies in order to avoid the cost of a burial (donated bodies are cremated and the cremains are returned to the family). However, since cadaveric organ donation disqualifies a body from being a candidate for whole-body donation, funeral regulations should also decrease organ donations.
In order to persuade families to be willing to donate organs from brain-dead relatives who could be a potential donor, some states have toyed with the idea of defraying the cost of burial, but none has followed through. For example, in 1999 Pennsylvania lawmakers suggested attempting to reverse the disincentive for organ donation by offering $300 to families of cadaveric organ donors—money they could use to defray the cost of burial. However, after many revisions and legislative setbacks, the law was never passed, partially because it was seen as a direct violation of the NOTA of 1984.
State funeral regulations take the form of either regulations on the licenses issued to individuals who wish to become funeral directors or regulations on establishments that wish to offer funeral services. The former regulations dictate the number of years and type of training required for becoming a funeral director. The most common requirement of this type is that individuals who wish to become funeral directors must be trained as embalmers. If states have this type of requirement, then they often also require additional years of training and do not offer licenses to those who wish solely to cremate (direct disposition). Regulations regarding the licensing of funeral homes often specify the types of equipment that must be on-site and the types of services that must be offered by the establishment. The most common form of this regulation is that funeral homes must have embalming preparation rooms. States that have this type of regulation often also prohibit crematories and mortuaries from existing on the same site or require that crematories be located in cemeteries (Harrington and Krynski, 2002).
For the purpose of the current analysis, we refer to states that require either funeral directors to be licensed embalmers or funeral homes to have embalming preparation rooms as being in stringently regulated states. These states have been shown to have (1) higher costs for funeral services (Harrington 2007); (2) relatively fewer people choosing cremation (Harrington and Krynski, 2002); and (3) higher rates of whole-body donations (Harrington and Sayre, 2007). Higher rates of whole-body donations in these more regulated states may reflect families wanting to avoid the relatively more expensive funeral costs, about $546 higher in 2000 (Harrington, 2007), by choosing whole-body donations.
It is this empirical relationship that is related to cadaveric organ donations. If a brain-dead individual is a potential organ donor, the family must make a decision whether or not to donate the organs, or donate the whole body. If the family decides for organ donation, then whole-body donation is not an option since many medical schools will not accept a body from which the organs have been removed (Lee, 1997). By opting out of whole-body donation, the family also opts out of a low-cost option for disposition since the medical school pays the cost of cremating the donated body, with the family being responsible for the cost of transporting the body to the donation facility and sometimes a nominal fee for the cremains.5
A. Descriptive Statistics and Sample
Harrington and Sayre (2007) show that for the states for which they could obtain information on whole-body donations, the highly regulated funeral states had rates of whole-body donation per 1,000 in 1990 and 2000 more than twice the rates of the less regulated states. Our hypothesis is that families may substitute whole-body donations for organ donations in highly regulated states; hence we would expect lower organ donation rates for counties in highly regulated funeral states.
Our sample includes 1,419 counties from the same 22 states used in Harrington and Sayre's study: Alabama, Arizona, Arkansas, Colorado, Delaware, Florida, Georgia, Idaho, Kentucky, Maine, Maryland, Michigan, Minnesota, Nebraska, Nevada, New Mexico, North Dakota, Texas, Utah, Vermont, Washington, and Wisconsin.6 Of these counties, 1,129 are in the 16 highly regulated states, while the remaining 290 come from the relatively unregulated states—Arizona, Colorado, Florida, Nevada, New Mexico, and Washington.
Table 1 shows the average deceased kidney and heart donation rates for these counties in 1990 and 2000, as well as their average rates by regulation status. Overall, the average kidney donation rate was approximately 3.63 per 1,000, while the heart donation rate was under 1 per 1,000. Unlike Harrington and Sayre's findings in terms of whole-body donations, we find no significant difference in organ donation rates by regulation status. In fact, our estimates suggest that the average donation rates are higher for both organs in counties in states with more stringent funeral regulations. But again, these differences are not statistically significant. The only “exception” to this is heart donation rates in 1990, but the difference is in the wrong direction.
| Kidney Donations | Heart Donations | |||
|---|---|---|---|---|
| 1990 | 2000 | 1990 | 2000 | |
| Overallb | 3.60 | 3.67 | 0.96 | 0.83 |
| (7.91) | (8.05) | (2.55) | (2.19) | |
| N | 1419 | 1419 | 1419 | 1419 |
| Funeral regulationsc | ||||
| Stringent | 3.67 | 3.89 | 1.07 | 0.90 |
| (7.72) | (8.70) | (2.73) | (2.15) | |
| N | 1129 | 1129 | 1129 | |
| Not-stringent | 3.42 | 3.20 | 0.70 | 0.69 |
| (8.33) | (6.46) | (1.95) | (2.28) | |
| N | 290 | 290 | 290 | 290 |
- aMeans are weighted by number of deaths in county and data is from 1,419 counties of a sample of 22 states.
- bMeans between years for the same organ are not significantly different.
- cMeans between counties with stringent versus not-stringent funeral regulations for a given organ are not significantly different. The “exception” is for heart donations in 1990, but the difference is in the wrong direction—counties with stringent funeral laws are expected to have fewer organ donations for transplant.
B. Regression Results
Although an examination of mean deceased donation rates by regulation status does not support our hypothesis that more stringent funeral regulations lead to fewer organ donations, we have not controlled for other factors that might affect organ donation. Table 2 presents our regression results with the log of total organ donations (separately for kidney and heart donations) as the dependent variable in 1990 and 2000. Our primary explanatory variable of interest is a dummy variable on state regulations, which equals 1 if the county is in a state with stringent funeral regulations.7 If stringent funeral regulations lead to lower organ donation rates, we should find the coefficient on this variable to be negative. In order to account for other factors that might affect the trade-off between whole-body donations and organ donations, we also include a variable on whether a state prohibits government officials from donating unclaimed bodies to medical schools and a county-level measure of the distance to the nearest body donation program.
| Kidney Donations | Heart Donations | |||
|---|---|---|---|---|
| 1990 | 2000 | 1990 | 2000 | |
| Stringent funeral regulations (1 = yes ) | 0.072 (.057) | 0.026 (.067) | 0.046 (.035) | 0.012 (.012) |
| Log nearest anatomy program (miles) | – 0.208 (.039)*** | – 0.231 (.046)*** | – 0.133 (.022)*** | – 0.157 (.027)*** |
| College (%) | 0.021 (.004)*** | 0.020 (.006)*** | 0.014 (.003)*** | 0.012 (.003)*** |
| Median income ($1,000) | – 0.032 (.007)*** | – 0.028 (.006)*** | – 0.023 (.005)*** | – 0.016 (.003)*** |
| Under 18 (%) | 0.015 (.008)** | 0.025 (.006)*** | 0.009 (.006) | 0.011 (.003)*** |
| Elderly (%) | 0.001 (.010) | 0.005 (.005) | 0.000 (.006) | 0.003 (.003) |
| Log accidental deaths | 0.042 (.020)* | 0.050 (.018)** | 0.022 (.011)* | 0.037 (.011)*** |
| Log cerebrovascular deaths | 0.012 (.018) | – 0.005 (.016) | 0.014 (.006)** | – 0.006 (.008) |
| Married (%) | – 0.014 (.005)** | – 0.010 (.005)* | – 0.006 (.004) | – 0.007 (.002)** |
| Female (%) | – 0.004 (.002)* | – 0.005 (.001)*** | – 0.002 (.001) | – 0.003 (.000)*** |
| Constant | 2.095 (.522)*** | 2.74 (.471)*** | 1.00 (.318)*** | 2.21 (.281)*** |
| Observations | 1419 | 1419 | 1419 | 1419 |
| R-Squared | 0.322 | 0.362 | 0.272 | 0.303 |
- Note: The parentheses contain robust standard errors corrected for clustering at the state level.
- *Indicates significant at 10%;
- **Indicates significant at 5%;
- ***Indicates significant at 1%. All regressions also included the following variables not listed in the table measured at the county level unless otherwise noted: log total deaths, median family income, percent poor, unemployment rate, percent non-Hispanic White, percent Hispanic, percent Black, percent native, state per capita health expenditure, percent Evangelical Protestant, percent Mainline Protestant, percent urban, percent Catholic, percent Jewish, and a dummy variable for whether the state cannot donate unclaimed bodies to anatomy programs.
We also include a number of variables that have been shown to be related to the rate of organ donation (see Abadie and Gay, 2006). State level per capita health expenditures from the Centers for Medicare and Medicaid Services (2009) are included to capture the technology intensity of the health care sector. County-level economic and demographic variables are included to capture other factors that are thought to affect the rate of organ donation.8 These include socio-demographic measures taken from the 1990 and 2000 U.S. Census (U.S. Census Bureau, 2009a), unemployment rates from the Bureau of Labor Statistics (2009a), measures of the total number of cerebrovascular and accidental deaths from the Centers for Disease Control and Prevention (2009) and measures of religious affiliation taken from the 1990 and 2000 Census of Religions (Association of Religious Data Archive, 2009).9
We find that several of the control variables are consistently estimated to have a significant impact on organ donations. Proximity of the county to the nearest body donation program is associated with higher rates of organ donation. This effect is likely due to the fact that body donation programs are located in cities that also have organ transplantation centers. For example, in the first two states in our sample (Alabama and Arkansas), the body donation centers are in the same cities (Little Rock and Birmingham) which have the only organ procurement organization in the state. Proximity to a procurement organization increasing donation rates is consistent with previous studies that show proximity affects willingness to donate (see Grubesic, 2000). A higher percent of the population with a college degree also is associated with higher rates of donation, which may again reflect awareness of options or perhaps a “taste” for donation by relatively more educated individuals.10 The younger the county, the higher the number of organ donations. However, the higher the percent female and the higher percent married, the lower the number of organ donations. Note that since female and youth measure population characteristics and not the characteristics of the deceased, it is hard to interpret these effects. However, according to data from the OPTN, men donate nearly 60% of all the cadaveric kidneys and 70% of all the cadaveric hearts. Also, it is worth noting that since cadaveric organ donations require consent from next of kin (often a spouse) that marriage rates and organ donation rates are negatively correlated. Also, the types of deaths in the counties are correlated with the donation rate. Nearly a quarter of the donated kidneys and a one-third of donated hearts came from people who died in motor vehicle accidents, so the positive correlation with the number of accidental deaths (including motor vehicle accidents) is expected. It is less clear why the cerebrovascular deaths are only correlated with organ donation in one model since strokes are a major circumstance of death that leads to cadaveric organ donation.
However, our particular interest is whether stringent funeral regulations, by increasing the probability of whole-body donations, may decrease organ donations. Consistent with the findings from the descriptive statistics, our regression results do not support the hypothesis that families that reside in counties subject to stringent funeral regulations are significantly less likely to donate their deceased relative's organs. It is also true that regulations that govern whether the government can donate unclaimed bodies to anatomy programs are not correlated with organ donations (results not reported in table).
III. STATE LAWS AND LIVE ORGAN DONATIONS
Although the NOTA of 1984 prohibits directly financially compensating individuals, whether living donors or next-of-kin of deceased donors, numerous states have passed laws that provide some compensation for living donors. These laws are of two general types. The first of these laws targets government employees, typically allowing full- or part-time state employees up to 30 days of paid leave due to the donation.11 In 2000, Wisconsin was the first state to pass this type of law, followed by Maryland later that same year.
The second law allows individuals to take up to a $10,000 tax deduction for costs incurred due to the organ donation. Again, Wisconsin, in 2004, was the first state to enact this law. The monetary value of this deduction depends on the individual's state marginal tax rate, which depends on the state, as well as one's income and filing status. Table 3 shows several scenarios for two states that have this law, Wisconsin and Minnesota.12 A single filer in Wisconsin with an adjusted gross income of $25,000 faces a relatively high state marginal tax rate because as one's income increases, the value of the person's state earned income tax credit decreases. However, Minnesota's tax system results in more wealthy individuals facing a higher marginal tax rate. Yet even though these examples show that the value of the tax deduction law varies, it is clearly not a lot of money (less than $1,500) to compensate one for donating a kidney.
| Single and State Adjusted Gross Income of $25,000 | Married, Two Kids and Adjusted Gross Income of $50,000 | Married, Two Kids and Adjusted Gross Income of $150,000 | |
|---|---|---|---|
| Minnesota | $535 | $535 | $705 |
| Wisconsin | $1,293 | $826 | $650 |
- Source: Lynn Quincy and the MPR Tax Simulation Model.
Table 4 shows every state that has passed either or both of these laws and when the law became effective. As of August 2007, 24 states had passed one or both laws. In this section, we examine whether these financial incentives, although minor, may affect living organ donations.
| State | State Employee 30 Days of Paid Leave (month/year) | $10,000 Tax Deduction (year) |
|---|---|---|
| Arkansas | 4/2003 | 2005 |
| California | 10/2002 | |
| Georgia | 5/2002 | 2005 |
| Hawaii | 7/2005 | |
| Idahob | 7/2006 | 2007 |
| Illinois | 1/2003 | |
| Iowa | 7/2003 | 2005 |
| Kansas | 4/2001 | |
| Maryland | 10/2000 | |
| Massachusetts | 1/2006 | 2008 |
| Minnesota | 6/2006 | 2005 |
| Mississippi | 7/2004 | 2006 |
| Missouri | 7/2001 | |
| New Mexico | 7/2007 | 2005 |
| New York | 11/2001 | 2006 |
| North Dakotac | 5/2005 | 2005 |
| Ohio | 7/2002 | |
| Oklahoma | 7/2002 | |
| South Carolina | 8/2002 | |
| Texas | 9/2003 | |
| Utah | 7/2002 | 2005 |
| Virginia | 4/2001 | 2007 |
| West Virginiac | 7/2005 | |
| Wisconsin | 5/2000 | 2004 |
- aStates not included in the table had not passed either law as of August, 2007.
- bState was not included in our analysis since information on donations was not available from the OPTN website.
- cState law allows for 20 days of paid leave.
A. Descriptive Statistics and Sample
Over time, both live and deceased kidney donations have increased. In 1988, the first year data was available from the OPTN website,13 nationally there were 1,817 live and 3,874 deceased kidney donations. Live donations steadily increased and in 2001 reached 6,038 and for the first time, outnumbered deceased donations, which were 5,638 for the year. For 4 yr (through 2004) live donations exceeded deceased kidney donations, perhaps leading some to wonder if this was an indication that kidney transplants would become more and more dependent on live donations. However, since 2004, deceased donations again surpassed live donations.
Since we are interested in the impact of state laws that came into effect in 2000 or later, we restrict our sample to years including and after 1999 (specifically 1999–2006). Because donor information was not available for 12 states,14 our empirical analysis is based on data for 38 states.
Figure 1 shows the mean living kidney donation rate per 1,000 adults and the total number of living kidney donations over the time period of our analysis. The rate of living kidney donations was calculated as total live donations divided by the adult state population,15 and is included in the analysis since one might expect the number, but not necessarily rate, of donations to change as population size changes.
Live Kidney Donation Rate and Total Live Kidney Donations, 1999–2006
Figure 1 suggests that neither overall donation rates nor the number of live donations have monotonically increased over this period and in fact peaked in 2004. Tests of significance indicate that none of the mean donation rates are significantly different from the following year's value. However, the 2006 donation rate is significantly higher than the 1999 rate, yet not significantly different from the 2000 donation rate. Although there have been significant changes in live kidney donations since 1988 (both statistically and in magnitude), these simple comparisons suggest that over these 8 yr, particularly in terms of donation rates, there has not been an important time trend.
However, our particular interest is to examine whether these two laws affect live donations. Therefore, Figure 2 shows what has happened to donation rates in states that have passed a law (experiment) compared to states that have not (control). This figure contains average donation rates for experiment states starting from 2 yr before they passed the law (t− 2) through 2 yr after they passed the law (t+2).16 It also shows what happened in states that did not pass a law during that time, using 2002 as the t = 0 yr. The graph clearly shows experiment states saw a greater increase in the donation rate than did control states, but most of the rise in donations occurred before the passing of the law.17 While experiment states have a lower donation rate earlier in the period, the donation rates of experiment and control states are nearly identical by the end.
Live Kidney Donation Rates in Experiment and Control States
B. Model
Although an examination of mean donation rates does not support the hypothesis that these state laws have led to an increase in the rate of live donations, we have not controlled for other factors that may affect live donations that could be obscuring the impact of the state laws.
To obtain information on state organ live donation laws, we searched states' legislative websites for existing laws, or laws that had been introduced. To verify the cases when information on a law could not be found, e-mails were sent to both an organ donor organization in the state and a librarian at a state government library. Replies to these e-mails almost uniformly verified our findings.19
We estimate six models using a variety of approaches to measure the impact of one or both of these laws. The first model uses a dummy variable (law) for whether the state had passed one or both laws to examine whether if a state simply having a law that “rewards” live donors is correlated with more live donations.
The next two models focus on the impact of the state leave laws. Model 2 includes a dummy variable for whether the state has a law allowing state employees to take paid leave for an organ donation. However, since the law only applies to specific types of employees, Model 3 replaces the dummy variable with the number of people in the state covered by the law.20 We also include a measure for the years the state leave law has been in effect to allow for the possibility that the longer the law has been in effect, the better known the law will be to potential donors, and hence increase live donations.
Models 4 and 5 focus on the tax deduction law. Like Model 2, Model 4 simply uses a dummy variable for whether the state allows individuals to take a tax deduction for costs incurred due to a live organ donation. Model 5 replaces the dummy variable with an estimate for the value of the tax deduction assuming the individual takes the entire $10,000 tax deduction and is subject to the maximum state marginal tax rate as estimated by the National Bureau of Economic Research TAXSIM model. Both of these models also control for the number of years the law has been in effect. Finally, Model 6 combines the law variables from Models 3 and 5. For any of these models, if these laws increase live donations, we should expect a significant positive coefficient on the law variables.
We control for (the natural log of) cadaveric kidney donations for two reasons. A greater number of cadaveric donations may reflect a greater propensity for individuals in the state to favor organ donations. In this case, the coefficient on cadaveric donations would be positive, reflecting the general population's relatively positive attitude toward organ donations. On the other hand, a greater number of cadaveric donations may reduce the need for live organ donations and therefore have a negative impact on the dependent variable.
For the vector of demographic controls (DC), we control for (natural log of) the size of the adult population, family size, percent married, percent elderly, percent White, education, and family income. To estimate these variables (except for adult population), we use the CPS March Supplements for 1999–2006, restricting the sample to individuals 18 yr and older and use the mean values for each state by year.
Although a live donor could come from another state, most donations are likely from individuals living in the same state. Therefore, we control for the potential supply of live donors, represented by the adult population in the state.
The number of people in the family and percent married in the state are included since historically most often live donors are family members. According to the 2007 OPTN/SRTR Annual Report, over 68% of live kidney donors between 1999 and 2006 were blood relatives of the recipient. Spouses represented approximately 12% of live kidney donors. Although not ideal, number of people in the family and percent married in the state are an attempt to capture the degree of family ties in the state. We expect the coefficients on both of these variables to be positive.
We control for the percent elderly and percent White in the state because live donors are much less likely to be elderly and more likely to be White. Only about 1% of donors between 1999 and 2006 were over 65, while Whites made up approximately 70% of live kidney donations (OPTN AR 2007). Therefore, we expect the coefficient on percent elderly to be negative and the coefficient on percent White to be positive.
We use five categories for highest grade attained: less than high school, high school (the omitted category), some college, bachelor's degree, and professional degree. Although we do not have strong priors on the impact of education on live donations, we hypothesize that more educated people may be more likely to be live donors due to their experiences with the medical community and knowledge about organ donation. Finally, we include (the log of) total family income measured in 2006 dollars as a control.
Table 5 provides summary statistics. In our sample, the mean number of live donations was 263 per year. Mississippi had the fewest live kidney donations (at zero for three of the 7 yr), California had the highest number of live donations in a year (743), and Minnesota had the highest rate of live donations per 1,000 adult population at 0.123. Of the states that had passed a leave law, 248,000 employees in the state were covered by the law on average—ranging from 53,000 in Kansas to 482,000 in California. On average, the estimate of the value of the tax deduction was $667 for states that had passed the tax deduction law. The table also highlights the variation in the remaining control variables between the states.
| Variable | Mean | Minimum | State | Maximum | State |
|---|---|---|---|---|---|
| Live donations | 263 | 0 | MS | 743 | CA |
| Live donation rate | .029 | .000 | MS | .123 | MN |
| # Employees covered by leave law* (in thousands) | 248 | 53 | KS | 482 | CA |
| Tax deduction value* (in $) | 667 | 493 | MS | 809 | MN |
| Adult population (in thousands) | 5,433 | 896 | HI | 26,924 | CA |
| Cadaveric donations | 147 | 16 | NM | 838 | CA |
| Cadaveric donation rate | .028 | 0.006 | FL | 0.082 | KS |
| Family income (2006 $) | 66,673 | 49,056 | AR | 93,459 | NJ |
| % Married | 60.6 | 54.5 | NV | 67.1 | IA |
| % White | 81.9 | 19.6 | HI | 97.0 | IA |
| % Elderly | 15.4 | 9.9 | UT | 22.5 | FL |
| Family size | 2.8 | 2.5 | OR | 3.5 | UT |
| Education variables (in %) | |||||
| Less than high school | 20.3 | 13.2 | MN | 28.0 | TX |
| High school | 31.1 | 21.4 | CA | 41.9 | PA |
| Some college | 26.1 | 18.5 | PA | 35.7 | UT |
| Bachelor degree | 15.2 | 9.6 | IN | 24.0 | CO |
| Professional degree | 7.3 | 3.8 | MS | 14.8 | MD |
- *Only includes states covered by law and for the years the law was in effect.
C. Regression Findings
As discussed above, our models control for state laws on living organ donations, cadaveric donations, and demographic controls. The models also include year and state dummies, and adjust the standard error for clustering at the state level.21 There are a total of 304 observations (8 yr for 38 states).
Table 6 shows the regression results for the six models. The findings are very consistent across the models. We find (1) no significant association between one or both of these laws and live kidney donations, (2) the number of cadaveric donations are not estimated to significantly affect live kidney donations, (3) a 1% increase in the state adult population is associated with about a 1.5% increase in live kidney donations, (3) states with higher average family income have lower rates of live donations, (4) a 1% increase in the percent married in the state is associated with about a 2.5% increase in live donations, and (5) none of the remaining demographic controls or education variables were found to ever be significant.
| Variablea | Model 1 | Model 2 | Model 3 | Model 4 | Model 5 | Model 6 | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| estimate | st. error | estimate | st. error | estimate | st. error | estimate | st. error | estimate | st. error | estimate | st. error | |
| Law dummy | 0.007 | 0.066 | ||||||||||
| State leave dummy | 0.024 | 0.076 | ||||||||||
| Employees covered by leave law | 0.007 | 0.014 | 0.007 | 0.014 | ||||||||
| State leave years | 0.001 | 0.025 | – 0.001 | 0.024 | 0.000 | 0.002 | ||||||
| Tax deduction dummy | – 0.191 | 0.143 | ||||||||||
| Tax deduction value (in $100s) | – 0.014 | 0.022 | – 0.015 | 0.023 | ||||||||
| Tax deduction years | 0.046 | 0.071 | – 0.006 | 0.077 | – 0.015 | 0.079 | ||||||
| Adult pop (ln) | 1.389** | 0.638 | 1.348* | 0.680 | 1.344* | 0.679 | 1.519** | 0.640 | 1.526** | 0.635 | 1.430** | 0.700 |
| Cadaveric donations (ln) | 0.030 | 0.097 | 0.029 | 0.100 | 0.029 | 0.100 | 0.042 | 0.090 | 0.040 | 0.090 | 0.034 | 0.094 |
| Family income (ln) | – 1.191* | 0.590 | – 1.190* | 0.598 | – 1.194* | 0.600 | – 1.073* | 0.564 | – 1.127* | 0.572 | – 1.114* | 0.576 |
| % Married | 2.412* | 1.381 | 2.456* | 1.411 | 2.477* | 1.416 | 2.500* | 1.394 | 2.547* | 1.411 | 2.679* | 1.390 |
| % White | 0.867 | 1.522 | 0.842 | 1.522 | 0.844 | 1.525 | 0.910 | 1.565 | 0.933 | 1.550 | 0.885 | 1.554 |
| % Elderly | – 0.352 | 1.980 | – 0.336 | 2.009 | – 0.357 | 2.009 | – 0.452 | 1.904 | – 0.431 | 1.923 | – 0.384 | 1.978 |
| Family size | – 0.008 | 0.292 | – 0.011 | 0.286 | – 0.012 | 0.284 | – 0.106 | 0.292 | – 0.084 | 0.292 | – 0.106 | 0.280 |
| R2 | 0.975 | 0.975 | 0.975 | 0.976 | 0.975 | 0.976 | ||||||
- aAll models include education controls, year and state dummies.
- *Significant at the 10% level;
- **Significant at the 5% level.
- Note: All regressions are adjusted for clustering at the state level and standard errors are adjusted for heteroskedasticity.
Although the tendency is to hope to find significant impacts, in this case particularly for our variables that should capture the impact of these laws, our results should not be too surprising considering we are trying to estimate what may impact individuals' willingness to become a live kidney donor. The two state laws that are the focus of this analysis do not provide large financial incentives, and it is not surprising that they do not significantly impact one's decision to donate a kidney. Although occasionally people donate a kidney to an unknown recipient, the vast majority of live donations are given to a relative or friend. Unfortunately, none of our independent variables can capture a person's ability and desire to donate an organ.
IV. CONCLUSION
In this paper we examine several laws that could affect the financial incentive to donate organs for transplant. First, we consider if more stringent funeral regulations reduce the supply of transplantable organs from deceased donors. Second, we look to see if recent state laws that provide some financial compensation to live organ donors increase the supply of transplantable organs. In neither case do we find evidence that these indirect financial incentives impact organ donations.
Why don't these financial incentives seem to be important? There are numerous possibilities. First, potential donors (or their next-of-kin) may not even be aware of the laws, and hence aware of the financial incentives to donate or not to donate. For instance, when obtaining information on state living organ donor laws, the authors contacted individuals in over 20 states. No one immediately knew whether the state had either living organ donor law. Second, it may be that financial incentives are simply not important, either for next-of-kin of deceased potential donors or potential living donors. One can imagine, particularly for live donors, that other factors are much more important in the decision to donate. Third, perhaps financial incentives could make a difference, but the level of these financial incentives is not great enough to encourage people to donate who were not already inclined to do so. This does not imply that we suggest eliminating or not passing laws that provide some financial compensation to live donors (or next-of-kin for deceased donors). These laws seem to be a minimal way to acknowledge or thank those who choose to be live donors. It also does not imply that a “market” for organs would not be viable, since the level of these incentives cannot provide useful guidance on how people may respond to much larger financial incentives that would likely occur in a market.
Although there does not seem to be a financial response to the live donation laws, the increase in live donation rates immediately prior to the passage of these laws suggest that they may have a non-financial impact. For instance, during the period leading up to the passage of these laws, there may be increased media exposure highlighting the need for organ transplants and in turn, increasing live organ donations. Regardless of the mechanism, the passage of one of these laws minimally may lead to an increase in donations or at least, not adversely impact live donations.
Our findings suggest that it will take much more than these types of laws to increase the supply of transplantable organs. To ascertain whether financial incentives can impact donations, ideally one would randomly offer individuals varying levels of financial compensation for organ donations and use this data to analyze the impact of financial incentives. Although we could argue this would be valuable to estimate the impact for potential deceased and living donors, politically neither may be possible. Perhaps someday there will be enough political support to allow this type of experiment. In the meantime, to increase the supply of transplantable organs, the United States will need to aggressively pursue other avenues.
