What's So Bad About a 50-50 Fundraiser?

Reader Melissa Belvadi writes in with a question about preferences on fundraiser incentives:

Here in northeastern Canada, there is a very popular form of local fundraising called the “50-50.” Basically it’s a raffle, where 50% of the total money collected is then randomly given to one of the donors (odds weighted proportionately by the donation size), and the other 50% goes to the original ’cause’ of the fundraising, whether it be a local homeless shelter, a recent victim of something, or whatever that is of interest to the local community.

This strikes me as an incredibly bad deal, but a bit complicated to explain why, as it contains two components:

  1. As a gamble: poor expected value. I am not sure how to calculate this, but from my experience in Las Vegas where slot machines boast being set to 97% return ratios, a gamble where 50% goes to the “house” seems unlikely to be a good EV.
  2. As a charitable donation: poor “program ratio” — at most, 50% of my donation will go to the “program” (charitable cause) – this is considered a very poor ratio in the philanthropic world where typically 60% is the bare minimum acceptable – the BBB requires 65%.

The complication is trying to combine these two perspectives – if both are poor choices by themselves, should they “boost” each other’s value as a choice because of the extra value offered by the other, or reinforce each other’s “poorness” and make it an even worse decision?

When confronted with one of these, my response is usually to refuse to participate in the 50-50 concept, but if the cause is at all reasonable, to ask if it’s possible for me to donate directly to the cause without having 50% of my donation going to some other donor. Sometimes they say yes, but sometimes they’re so thrown off by my objections to the format, that they can’t deal with my counteroffer.

Is this just another case of “lottery is a tax on those who can’t do math,” or is there something more interesting going on here?

What are your thoughts on the 50-50 concept? Keep in mind what John List has taught us all about the tricks that make donors donate. In my view, there’s nothing at all wrong with the 50-50. I can see how some people, like Melissa, might rule it out. But I think a lot depends on how it’s marketed. Yes, as pure fund-raiser it’s inefficient; yes, as a gamble it’s inefficient; but as a hybrid, what’s not to like?


I've never seen such a thing promoted as a 50/50 fundraiser -- around here they are always called 50/50 raffle, and your impetus to pay is for the opportunity to win half the pot, not because you care where the other half of the money is going. They used to have one at pretty much every single high school sports game I ever went to, so I assume they were run by the boosters, but to be honest I never bothered to find out and I doubt most people did!

It wasn't something I ever participated in -- I never win and I certainly had stuff I wanted to actually spend that dollar on instead.


Most state lotteries in the U.S. that I have heard of have a return of 50 cents on the dollar, or the same as a 50-50 raffle. The difference in this is that the beneficiary is a defined charity or organization, while with the state lotteries it just goes to the state. To me, a 50-50 would be a better choice for anyone willing to purchase a scratch ticket because you determine who benefits from your donation, and you get the gambling aspect to fulfill your need to gamble. The difference is that these raffles are honest about the payouts (it is in the name), while normal scratch tickets blind you with silly games and potential huge payouts. Whether or not it is beneficial as a way to get donations would have to be determined by a psychologist and several studies, but I would think that it would be a way to trick people into donating who would not be willing to drop money in a jar.


I think there another aspect being overlooked here: effort required by the charity/organization holding the raffle. It takes quite a bit of work to gather raffle prizes and often the amount raised is out of balance with the value of the prizes. If there are good prizes, you're better off doing a silent auction. A 50/50 can be run with no more planning than is needed to pick up a spool of raffle tickets.


We have 50-50 fundraisers at my kids' school (along with several other). I've never won, but it doesn't stop me from entering. And I've been told that proper etiquette for the fundraiser if you win, is to decline the winnings and donate it back to the organization. In this case, it is my children's school, so even if a parent (assuming it's mainly parents who participate) declines the money, she is still benfitting because it benefits the school which benefits her child.

Perhaps, since it's a private, catholic school, there is a guilt factor involved. The fundraiser occurs at triva night and the nuns roam around and ask for donations-how can you say no? Plus, the winner is announced at the end of the night, so everyone there can see what the winner does.

Enter your name...

It always disturbs me to see a Catholic organization promoting gambling. Their official doctrinal statements condemn gambling as a morally risky (but not, I believe, always or automatically sinful) behavior.

It seems to me that a church should not directly engage in behaviors that they believe can be harmful or sinful. I believe that famous line runs, "Lead us not into temptation", rather than "Let's see just how close we can get to the edge of the cliff before someone gets hurt."

Mike B

50/50's make sense where the potential pot is small enough so that the potential house take wouldn't justify the hassle to run the event in the first place. To the player the charitable donation comes in the form of the decreased payout.

My mother was actually a pretty big 50/50 innovator on the local swim meet circuit. Normally the prices had been $1 for 1 ticket and 5 for $3. She introduced an arm's length for $5 (the person could choose whose arm they wanted even) which had the effect of bumping everyone up to the $5 price point and also making them feel that they were getting better value for money (after all they had a whole roll of tickets instead of 1 or 5). She was able to double or triple the standard 50/50 take and the "arms length" option became standard throughout the area in the following years.


There is nothing at all wrong with it- it's genius, in fact. It is the perfect mathematical intersection between self-interest and philanthropic interest. Contrary to Ms. Belvadi's assertion that "both ways you lose", I'd say either way you win: you have a chance to win the money, you get the entertainment value of participating in the game, and you get to feel good knowing your gaming is benefiting a charity.
If you're still skeptical, here's the salient question: is this method of fundraising more effective for the charity than simply soliciting "pure" donations? I'd wager the answer is a resounding "yes"- and thus the idea of rejecting it as not an effective donation to the charity is misguided.



People go to Vegas thinking they will get lucky and win money, people go in on a 50/50 with the main intention of donating to the high school band, or whoever is running it, and maybe a slight hope of winning 50%. It's more entertaining and less obnoxious than "please donate to our cause".


50-50 draws are by design easy to administer since they amount to purchasing a ticket roll at a dollar store, organizing some people to sell tickets, having the draw and distributing half of the collected money. No one cares about the odds or payoff for the charity/team, transparency and administrative simplicity are what matters when someone decides to hold a 50-50 draw.

Bearman Cartoons

Jim for what the boy scouts charge for that popcorn they could buy it retail, double the price an sell it. The customer gets a better deal and the scouts get 50%

I am more likely to buy cookies from the girls multiple times because of the lower cost. Boy scouts I may buy once

Enter your name...

These are theoretically illegal in California, where charity raffles must pay out no more than 10% of the proceeds.

They are also common. You just do a little accounting sleight-of-hand and declare that the 50% prize doesn't actually come from the raffle proceeds, but from some unrelated prior donation. This is presumably how the Yerba Buena Center for the Arts in San Francisco is justifying their "Dream House" raffle, which offers a $3 million prize with a maximum potential revenue of $6 million. In another proof that voter initiatives and amendments are among the worst written laws on the books, there's no limit to the actual payout, but only to the part of the receipts that are paid out. According to the law that authorized charity raffles, you could legally take a million dollars of non-raffle-related donations, sell a single $1 raffle ticket to each of the organization's board members, and raffle off the million-dollar prize. So long as each of those $1 ticket prices ended up in the charity's bank account, that would be considered a raffle for which 100% of the proceeds benefited the charity.



I think the best way to do the math is to separate your contibution. Lets assume you pay $20 to enter the 50/50. $10 is a donation with 100% efficiency and $10 is a gamble with $10 expected value. Both are very good outcomes.


To make it totally efficient, you should get a tax receipt for the $10 donation.


The 50% isn't wasted--it incentivizes others to donate.


Starting with benefits to the charity. That's easy I think. 50% of something is better than 100% of nothing. So if they get a donation through this 'lure' then all credit to them. They weren't getting anything in the first place.

But for a pure gambler...the problem is information asymmetry. Specifically what the total prize pool is so that they can work out their return on investment for a given bet. As I understand no-one knows what the total is until the draw.

I'm a bit sketchy on the math and probabilities here, but lets start with the notion that the prize pool is known to be $100. A $50 investment gives the gambler a 1 in 2 chance of winning their money back. Well no one is going to go for that. The other side of that is they have a 1 in 2 chance of losing their money. But if they know the total prize pool they can make a decision on the risk and reward they will accept. What about a $30 'bet' in a prize pool of $100. Now the chances of winning is about 1 in 3, but the ROI is 67%. Is that an acceptable gamble. Maybe it is.

But what if the prize pool was increased to $200 without the gambler knowing. Now the odds are 1 in 4 chance of winning their money back (albeit doubling it). If that were the odds in the previous example, they would have only had to 'bet' $25, saving themselves $25.

So in summary, the thing that I think makes this a bad (meaning uniformed) gamble is not knowing what the total prize pool is so the gambler can work out the chance of winning and ROI. And just to add, the draw is presumable randomly drawn. Not much different to roulette, but at least in roulette you know the odds you are dealing with.



Isn't this a theory paper by John Morgan in the Review of Economic Studies: "Financing Public Goods by Means of Lotteries", 2000? Bottom line: it balances the trade-0ff to contributing too little to a public good on the one hand, and buying too many lottery tickets as it increases your chances of winning to the detriment of others. Some suggestions of commenters here were along the same lines, I guess.