Modeling Donor Decision Making

I recently wrote a relatively popular post where I argued that there is a need to look at philanthropy through the lens of behavioral economics.  Before we can can really dig into the behavioral considerations, we should start by thinking through how one would develop a rational model of donor decision making.

Economists would have the equation look something like this: Max u(x) where x is money and the function u() represents the donor’s utility function.  The current discourse (I am referring to the randomista’s model of efficient aid allocation) further assumes that u(x) = f(x) where f() maps money to social welfare.  The critical assumption here is that the objective of the donor is maximizing social welfare.

Even if we assume perfect rationality, this is an over-simplification for sophisticated foundations let alone individual donors.  In reality, the functions for donor utility and social welfare are not the same.  Not by a long shot.  Impact maximization is typically bounded by donor affinity for a particular cause.  Frankly this makes sense, because the organization needs to build sufficient expertise in a domain to evaluate giving opportunities.  (Side note: GiveWell has been digging into this lately, see their excellent blog post on strategic cause selection).  So the equation would look more like Max u(x) where u(x) = f(x) such that f(x) is confined to the opportunity set of cause y.  In constraining by cause, donors are searching for local maxima, not the highest possible social return on investment.

The other problematic assumption with the above models is that donors can assess all possible giving opportunities to determine the highest impact ones.  This is obviously infeasible, but we can imagine at least choosing those that are the most promising among a set of opportunities.

While this might seem like a reasonable way to model the decisions of large donors, it turns out this equation makes zero sense for individual donors.  Hope Consulting recently examined the market for charitable donations (see their must-read report, Money for Good).  They surveyed households with incomes over $80k, which represent 75% of charitable donations from individuals (approximately $150B in 2010).  Their research revealed that only 35% of individual donors do research and when they do, it’s to validate an organization rather than compare giving opportunities.  Furthermore of the 35% that does research, 75% spend less than 2 hours researching and almost always go to the organization itself to collect information.  Ultimately the report concludes that only 3% of people donate based on the relative performance of an organization.

So let’s be frank. For all but a tiny fraction of individuals, maximizing impact isn’t part of the equation at all.  The social welfare created for individuals donors looks more like the sum of f(x) where f(x) is a particular organization’s impact, and x > 0 if some other criteria is satisfied.  The correct model for individual donor behavior lies in understanding when x > 0.  The implication here is that bringing these two equations into closer alignment could yield big gains in social welfare.

How then, could we model individual donor decisions?  I would approach it by assuming there is some “hurdle” that must be met in order for an individual donor to give.  Furthermore, I would assume that the level of the hurdle is not constant, but varies over time based on factors such as whether the donor recently made a large contribution elsewhere or whether the donor just received a windfall.  What factors would go into determining if a giving opportunity meets the bar?  They’d obviously be weighted differently based on individual preference, but I’d want to include the following:

  • Relationship with the organization (it’s my alma mater/church)
  • Connections to the employees of the organization (my friend runs it)
  • Experience with the cause (my grandmother suffered from Alzheimer’s)
  • Proximity to the donor (I want to help my community)
  • Relationships to others supporting the cause (She’s running a 10k to support breast cancer)
  • Chance (a volunteer stopped my on the street)
  • Tangibility (I can help this entrepreneur in Bogota get a loan)

This last piece, tangibility, is so important it deserves its own post.  This one is long enough as it is, thanks for bearing with me if you’ve made it this far.

One response to Modeling Donor Decision Making

  1. Great beat ! I would like to apprentice while you amend your web site, how can i subscribe for a blog site?
    The account aided me a acceptable deal. I had been a little bit acquainted of this your broadcast offered bright clear idea

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s