by George Taniwaki
There is an ongoing argument regarding whether we as a society should pay people to donate a kidney. These arguments, both pro and con, revolve around two issues, whether such payments are the right thing to do (ethics) and whether they would increase the number of available organs (economics). This blog post will describe the economic effects of payments.
Before analyzing the effect of payments on the supply of donors, I want to assure readers that payments can be regulated. For instance, nearly all the blood, plasma, and platelets in the U.S. is collected from unpaid donors. Yet at the same time, there is also an active government regulated market for plasma. Similarly, family members and friends are a common source of donor eggs, donor sperm, and surrogates to allow individuals to have a child. But there is an active market for these as well.
An organized market for donor organs would not likely include person-to-person transactions. Rather, it would involve highly regulated, non-profit entities that would act as intermediaries between donors and patients, similar to the existing network of organ procurement organization (OPO) that recover and distribute organs from deceased donors. In other words, ignore the image in Figure 1.
Figure 1. Kidney for Sale t-shirt. Image from zazzle.com
One of the arguments against paying donors for organs is that it will favor wealthy patients who can afford the price. That is not necessarily so. Laws can still be written to prohibit individuals or hospitals from making payments to donors. The payments can be regulated to only allow insurance companies and other government sanctioned groups to make payments. Similarly, the organs collected from donors need not be transplanted to patients based on ability to pay for the organ. They can be allocated by whatever method is deemed medically and ethically justified.
More patients could benefit from transplants
Many kidney disease researchers, ethicists, and economist agree that under the right circumstances, increasing transplant rates would be a good thing. First, transplants improve medical outcomes. Second, transplants save money.
Studies have shown that patients with end-stage renal disease (ESRD) who receive transplant therapy live longer than those who receive dialysis therapy (U.S. Renal Data System 2013 Report). This is true even after adjusting for the fact that transplant patients are healthier on average than the overall kidney patient population (R. Wolfe, et al., New Engl J Med Dec 1999).
The data also shows patients who receive transplant therapy report a better quality of life than those who receive dialysis therapy (W. Fiebiger, et al., Health and Qual Life Outcomes Feb 2004).
More transplants would save money
In addition to being better for the patient, transplants can save money. Dialysis therapy costs about $75,000 per year per patient. Transplant therapy costs about $150,000 for the first year (evaluation, surgery, recovery, and follow-up) and then $15,000 per year thereafter (antirejection medication, infection control, and monitoring). Over the lifetime of the graft, a living unrelated donor can save society $94,000 compared to dialysis (A.J. Matas and M. Schnitzler Amer J Transpl Feb 2004). Adding the value of the additional 3.5 quality-adjusted life years for the patient increases the social benefit to $269,000.
A recent paper by B. Manns et al. (Clin J Amer Soc Nephr Dec 2013) indicates that even a 5% increase in the number of donors would justify a payment of $10,000 each by providing an incremental cost-savings of $340 and a gain of 0.11 quality-adjusted life years.
There is a shortage of suitable organs
The reasons more kidney patients don’t pursue and receive transplant therapy are not fully understood. One thing is certain though. The number of viable organs that become available each year is significantly lower than the number of patients newly diagnosed with ESRD. Thus, the expected wait time for a transplant continues to get longer (up to 8 years in California).
About 15% of patients on the waiting list die each year, so the proportion of patients who die without ever getting a transplant increases as the wait gets longer (over 50% in California). This long wait may deter some patients (and their doctors) from even starting the transplant evaluation process. As of this writing, there are 98,935 people in the U.S. waiting for a kidney transplant.
According to the U.S. Renal Data System, there were 115,643 people newly diagnosed with ERSD in 2011, the latest year data is available. This includes 2,855 who received a preemptive transplant (meaning they received a transplant before having to go on dialysis). In contrast, the Organ Procurement and Transplantation Network (OPTN), shows there were only 16,814 transplants performed in the U.S. in 2011. The breakdown by donor type is shown in the table below.
|Living directed donor||3,761|
|Living exchange donor||575|
|Living nondirected donor||157|
|Deceased directed donor||123*|
|Deceased nondirected donor||12,198|
*Assumes that 1% of deceased donor transplants are directed (OPTN 2009)
Of the total, 3,761 came from living directed donors, meaning the donor and the recipient knew each other. 575 came from exchange donors, meaning the donor knew the intended recipient but was incompatible so donated to a stranger who was in the same position and they swapped kidneys (for details see Mar 2010 blog post). 157 came from living anonymous or nondirected donors, meaning the donor did not have an intended recipient (similar to most blood donations). Finally, 12,321 came from deceased donors (of which all but about 123 are nondirected).
Costs to becoming a live kidney donor are high
For now, we will ignore the impact of paying for deceased donor organs and focus on a possible market for live donors. Further, we will ignore the ethics and legality of paying people to become live kidney donors. We will cover these issues in a future blog post. For now, we will explore the economics of paying for live donors.
Being a living donor can be expensive. The evaluation and surgery are paid for by the recipient’s insurance. However, there are lots of out-of-pocket costs such as travel to and from the transplant hospital for evaluation. In some cases, there can be multiple trips and may require a hotel stay for out-of-town donors. There are also opportunity costs, such as lost wages (or foregone billings for the self-employed) for the time spent in evaluation, surgery, and recovery. The time spent at home after surgery can vary from a few days to over a month, so this is a real burden for people who don’t receive sick pay or disability insurance from an employer. I estimate the total out-of-pocket and opportunity costs for a typical donor to be about $2000.
Usually, all of these costs are borne by the donor, meaning most donors are wealthy. Sometimes, the recipient will pick up some of these costs, especially if they are wealthy. Sometimes the donor and recipient conduct a fund-raiser to pay these costs. Finally, there are several charities that provide reimbursement if the donor or the recipient cannot afford the financial burden of paying for a living donor transplant. The best known of these is the National Living Donor Assistance Center.
Supply curves for nonaltruistic, nondirected donors
To analyze the effect of payments for kidney donors, we will use the basic technique used by economists called a supply curve. The supply curve shows the quantity (Q) of organs supplied for any price (P). We will look at the impact of paying for kidneys on three groups of living donors.
The first group is the nonaltruistic, nondirected (NAND) donors. This consists of the population of people who are aware of the existence of people who need a kidney transplant but don’t know anyone personally who needs a kidney. Further, they may be willing to donate a kidney, but have no desire to donate a kidney for altruistic reasons.
Figure 2 shows a hypothetical supply curve for kidneys from this population. At the current offering price today (Pcur), the quantity of kidneys offered by NAND donors is zero. Note that Pcur is negative and reflects the costs associated with being a donor.
Raising the offer price will not result in any donors appearing until an offer of PNANDmin is made and the first donor will step forward. This initial price may be quite high due to what is called the repugnance factor by economist Alvin Roth (J Econ Perspectives, Summer 2007). (Repugnance will be discussed again when we explore the moral and legal issues surrounding payments to donors.)
As the price rises, more donors appear. However, at some point there may be some proportion of these potential donors who will be very reluctant to volunteer, regardless of the amount of money offered (perhaps because of very high repugnance, fear, or dislike of pain). At this point the supply curve will rise steeply, until you reach the last person in the population (QNANDmax) where a very large sum of money must be offered before they will be willing to undergo kidney donor surgery.
Figure 2. Supply curve for nonaltruistic, nondirected donors
Note that I made a simplification in the supply curves shown above and below. I assume the out-of-pocket costs and opportunity costs for all donors is the same and equal to Pcur. Actually, this is not true and these costs can vary widely. However, allowing for varying costs makes the analysis much more complex without adding any new insights.
As an aside, behavioral research shows that people’s preferences are not stable, called the endowment effect. For instance, many people may say they would not donate a kidney for $20,000. But imagine what would happen if we gave those people the $20,000 first and ask them to consider what they could do with that money. Then we wait a few minutes and ask them if they would rather give the money back or donate a kidney. At that point, many may decide donating the kidney is their preferred choice.
Supply curve for directed donors
The second group we want to look at is potential directed donors. These are people who know someone who needs a kidney transplant and may be willing to donate to that person. The reasons may be altruistic, self-interest (not wanting to lose a relative or friend), or perhaps even coercion by the recipient or family members. Regardless of the reason, we can draw a supply curve like the one shown in Figure 3.
This curve looks very similar to the one in Figure 2 except it is shifted down. That is, once a potential donor develops a connection to the recipient, the minimum reservation price drops. That’s because the act of donation generates utility for the donor. At the current price of Pcur there are QDDcur donors.
Figure 3. Supply curve for directed donors
Raising the price offered to this group should increase supply, even if the offered price is below PNANDmin. Just reimbursing every donor’s out-of-pocket and opportunity costs could have a significant impact on supply. However, the supply is limited to QDDmax based on the total number of people who know someone who needs a transplant.
Supply curve for altruistic nondirected donors
The third group we want to look at is altruistic nondirected (AND) donors. Even though these donors do not know the recipient, and in fact often will never know the recipient, the supply curve for this group looks very similar to that of the directed donors. The utility an AND donor derives from her donation is not from helping a known person. Perhaps, it comes from imagining that the donation is helping a deserving person, or helping society as a whole, or the donation represents an act of altruistic sacrifice. At the current price of Pcur there are QANDcur donors.
Similar to the case for directed donors, just reimbursing every donor’s out-of-pocket and opportunity costs could have a significant impact on supply. Offering a payment (which a truly altruistic donor could decline and donate to charity) may increase the supply as well. However, it is not likely to have a large effect. I suspect the supply of altruistic donors is inelastic. I also believe the total number of people who would be willing to donate to a stranger QANDmax is limited as well, though probably significantly larger than the current 150 per year.
Figure 4. Supply curve for altruistic, nondirected donors
Shifting the supply curve
There is an alternative response to raising the offering price to AND donors. Since the utility the AND donor receives is dependent on psychological reward, any action that reduces the value of that reward may shift the supply curve upward. At the limit, the AND donors will become a NAND donors. In the worst case, the former AND donors may have a higher reservation price than the NAND donors causing the supply curve to be above the curve for the NAND donors (dashed brown curve S’ in Figure 4).
If this supply curve shift occurs, then paying donors could have the perverse effect of reducing the total number of donors until price PNANDmin is exceeded and NAND donors begin to appear.
Conversely, a well-crafted marketing effort to encourage more people to become AND donors can keep the AND curve from shifting upwards. It can also convince some NAND donors to reconsider their position and become AND donors, causing the total number of NAND donors to shrink and the number of AND donors to rise (shifting QNANDmax to the left and QANDmax to the right).
Add it all up
Combining the three supply curves would create an overall supply curve that would look similar to the solid line in Figure 5. At the current price Pcur, the number of donors is QDD+ANDcur. When the price reaches PNANDmin, the NAND donors will begin to enter the market.
If making payments causes all the AND donors to become NAND donors, then the supply curve shifts to upward as shown by the dashed line S’. At the current price Pcur, the number of donors falls to QDDcur. When the price reaches PNANDmin, the NAND donors will begin to enter the market. When the price reaches P’ANDmin, the former AND donors will enter the market. Note that even if all the AND donors become NAND donors, there will still be a market price somewhere above PNANDmin that will result in more donors than are currently available at the current price of Pcur.
Figure 5. Cumulative supply curve for all donors
For more on the economic analysis of organ markets, see the following papers.
A. Tabarrok. Library Econ Liberty, Aug 2009. Discusses payment for deceased donor organs.
Scott Halpern, et al. Annals of Int Med, Mar 2010. 342 participants were asked whether they would donate a kidney with varying payments of $0, $10,000 and $100,000. The possibility of payments nearly doubled the number of participants in the study who said they would donate a kidney to a stranger. Payment did not influence those with low income levels more than those with high incomes.
Gary Becker and J.J. Elias. J. Econ Perspectives, Summer 2007. A thorough analysis of the cost and number of transplant performed if payments were allowed for the donation of both live and deceased donor kidneys. It also counters the arguments against payments.