After I closed my last company I found myself growing increasingly pessimistic. Pessimistic that my career to that point had generated any real social impact. Pessimistic that social programs had any real effect. I had figured out how to make a living in the social sector, but not social change.
In the year or so between shutting my business down and joining the Family Independence Initiative (FII) I pursued a number of opportunities. At one point I found myself listening to a coworker philosophize about how her intense optimism fueled her quest to end poverty in Sub-Saharan Africa. I remember thinking to myself how I wished I shared her optimism, and lamenting my loss of faith in social sector work. Appropriately, this lecture on optimism was delivered in a luxury vehicle with heated leather seats on our way to discuss a six-figure Word document prepared for a multi-billion dollar foundation. Africa has never been so saved.
Also during this period before joining FII I was courted by a social sector technology startup. My experience at the intersection of technology and nonprofits was attractive. However, the CEO of this company made it clear that my lack of faith in the social sector was not, telling me I was “too pessimistic” for the job.
For a long while I thought she was right. Why couldn’t I just believe in the social sector, and enjoy the cognitive dissonance of doing well for myself while believing I’m doing well for others as well? As much as I wanted to believe, I just didn’t.
It wasn’t until I joined FII that I realized what true optimism is, and how in my opinion the social sector masks extreme pessimism as false optimism.
At the heart of any social intervention is the theory of change. The theory of change assumes that some form of external intervention (tutoring program, job training, etc.) can drive a positive outcome (kids graduate from school, parents get jobs).
In human services, the theory of change is often steeped in a deep rooted belief that the poor cannot achieve on their own. These stereotypes include wrongheaded beliefs that the poor are:
- Bad parents
- Don’t value education
- Are not capable of holding jobs
- Don’t care for their health
These assumptions are not only wrong, but they underscore a cynicism that undermines the presumption of philanthropic optimism. Poor people can and do succeed everyday on their own terms without any external “social intervention”. We see this at FII and GiveDirectly is making this case in international development.
An optimist all along
I used to think I was a formerly optimistic person who had become a pessimist after realizing the social sector did not have the impact I had hoped. The exact opposite is true. I started my career with a pessimistic view of the poor. I believed they could not succeed. I was wrong, I was a pessimist.
Now I know the poor can and do succeed every day. I am an optimist. But the social sector is not, and with good reason. The optimism I speak of undermines the fabric of the social sector. Trust in people ironically is an affront to those of us who make our living “aiding” the poor. If they can do it on their own, what do they need us for?
Indeed, what do they need us for.
The social sector business model, of which I’m no doubt a part, is to sell donors on a disbelief in the poor, and then to sell their outcomes as ours. But nonprofits do not create social impact, people and families do.
Optimism is believing in people. Optimism is investing in families. Anything else is pessimism, although I’m not terribly optimistic the social sector will change.
It wasn’t long ago that data existed at the outskirts of the nonprofit psyche. In the last few years however, interest in social sector data has spiked, even if capacity has not kept pace.
A few pioneering nonprofits are making big bets on the transformative potential of data by hiring Chief Data Officers, like my employer the Family Independence Initiative (FII). For data to truly be transformative however, it cannot be the exclusive domain of those with “data” somewhere in their job descriptions.
Fundamentally, data analytics is about listening in aggregate, listening to those you serve, about what is working, and what is not. Everyone in your organization needs to listen. Therefore, everyone in your organization needs access to data.
When I joined FII there was a clear line between the “data people”, who had access to our analytics, and everyone else. Like most nonprofits, when someone outside the data team needed to access the data, a request would come in to the data team and someone would perform some analysis and report back.
When I joined FII one of my initial focuses was making sure everyone in the organization had access to our data. To achieve data access ubiquity, I built a series of internal facing data dashboards that empower staff to explore and extract the data they need via a web-browser on their own.
Shiny dashboards to the rescue
I conduct my data analytics with the popular statistical programming language R. R provides tremendous power to split and explore data as well as build any range of models from econometrics through machine-learning.
In order to get the power of the R language into the browser I developed FII’s data dashboards using a server framework called Shiny Server. I taught a surprisingly well attended session at this year’s Do Good Data conference on building Shiny dashboards, and am open to writing more technical posts on using Shiny in nonprofits if there’s interest.
Shiny has been a godsend, enabling me to spin up new dashboards practically at will. Our data dashboards are organized by topic. For example, we have dashboards for demographics, financials, education, health, etc.
Each dashboard allows staff to input custom report parameters, such as the time interval, specific demographics characteristics, and variables the staff member wants to explore. Below is a screen shot of our financials dashboard.
The beauty of Shiny is that you have the full power of R underneath, so your dashboards are only limited by your imagination (and of course R skills). Below is another example of one of our dashboards, this one showing educational outcomes by age ranges:
I’m heartened that the social sector is recognizing that data is important, but now begins the more difficult work of putting that realization to practice. The first step I believe is getting data out of the “data dungeon” and into the hands of everyone in the organization. After all, if data analysis is really listening in aggregate, who in your organization wouldn’t be aided by listening more closely?
Last week I saw an article from Time titled “Pediatricians Should ‘Screen’ Kids for Poverty, Says Group” making the rounds in my LinkedIn network. The article is about an American Academy of Pediatrics recommendation that doctors be responsive to how poverty can affect health by referring low-income families to social programs. The recommendation is based on the largely accepted concept of the social determinants of health.
The World Health Organization defines social determinants of health as “…the conditions in which people are born, grow, work, live, and age, and the wider set of forces and systems shaping the conditions of daily life.” The literature connecting poverty and health is solid. As the Center for Disease Control states, “We know that poverty limits access to healthy foods and safe neighborhoods and that more education is a predictor of better health.” This is both intuitive and factual.
I have no qualms with the linking of poverty and health outcomes, but rather object to the suggestion that one ought to be diagnosed poor, as this perpetuates a stereotype that poverty is something that is wrong with the poor rather than the result of a social and political environment we (humanity) collectively create.
Furthermore, the analogy wrongly simplifies poverty into something that is curable via proper prescription. From the Time’s article:
The recommendation also provides guidelines to help pediatricians connect families who might be struggling to the proper resources, from local housing bureaus to food pantries and job listings. The hope, says Dreyer, is to help the 50% of families who currently qualify for additional support but aren’t getting it to access the resources they are entitled to.
The recommendation makes it sound as though once diagnosed, poverty is “cured” via the social safety net. Implicit in this analogy is that the cure for poverty is proper prescription and utilization of the right dosage of social service intervention. In this thinking, either the poor do not know about or refuse to utilize services that are assumed beneficial.
Given this logic, the obvious question is why do so many families not pursue such help if the safety net is the cure to poverty? The core assumption in this line of reasoning is that families do not know about available services. Indeed I made this assumption for a number of years while selling systems to governments and nonprofits that “connected” people to social programs.
Since joining the Family Independence Initiative and getting access to a sizable volume of data directly from families moving in and out of poverty, my perspective has shifted significantly. Whereas I used to assume people simply did not know about social programs, I now wonder if utilization is low not on account of informational asymmetry, but rather because:
- Existing services do not necessarily help as much as we assume.
- The social stigma of accessing social programs outweighs the benefit.
- Services are such a small part of escaping poverty as to not be worth the hassle.
The poor are not sick, even if they have worse health outcomes, as poverty itself is not a disease. It is the absence of money, a social phenomenon we collectively support, and often benefit from. The causes of poverty are not individually acute, but rather socially, racially, and spatially broad.
Indeed, the disparity in life expectancy between lower and upper income households is growing. Where you live and how much you make can affect your health in lots of ways, from experiencing higher levels of violence, air pollution, poor food access, and several other factors. None of these are solvable through a diagnosis of poverty.
The real “cure” to poverty is not in small doses of guided interventions at all. Nonprofits are not doctors of poverty. As sure as evidence supports the social determinants of the health, a growing body of literature also supports the benefit of investing directly in low-income families.
Analogizing the poor as diseased perpetuates the myth that the poor are weak, feeble, and infested. Nothing could be further from the truth.
There is a subtle shift taking place in pockets of the social sector, challenging the historically interventionist approach to social change. Microfinance was perhaps the first in modern memory (or at least my memory) that eschewed the conventional wisdom that nonprofits know the path out of poverty better than the poor themselves, extending loans to low-income families to use as they saw fit.
More recently GiveDirectly has popularized the concept of unconditional cash transfers, whereby cash is given to low-income families with no strings attached. Domestically, homeless services are increasingly realizing the futility of trying to “treat homelessness” while people live on the streets, and instead are shifting resources toward simply putting people into homes.
Similarly, at the Family Independence Initiative we are investing in the initiatives low-income families are taking across the United States to improve their own lives and their communities. What all of these approaches have in common is that none of them make any assumptions or judgements about who the poor or homeless are. There is no real theory of change in the traditional sense. No layering of one expected outcome that ought to come before another.
Instead, these are straight forward common sense approaches that put trust in the poor and make them the center of social change. If someone needs a loan, let’s give them one. If someone is extremely impoverished, let’s transfer cash. If an individual is experiencing chronic homelessness, extend that person permanent housing.
All of these approaches are inherently anti-interventionist, and they seem to work really well. So why don’t we see a tidal wave of anti-interventionism?
The anti-interventionist threat
While anti-interventionism is great for the poor, it’s an affront to much of the nonprofit sector. Nonprofits raise funds on the assumption that their programs and services hold the key to lifting low-income families out of poverty.
These anti-interventionist approaches greatly reduce the role of nonprofits, as in the anti-interventionist’s view the nonprofit is no longer the driver of change, but the impoverished themselves. In a world of anti-interventionists, nonprofits are reduced to distributors of funds rather than architects of change.
In the interest of self-preservation, nonprofits have and will continue to argue against anti-interventionism. However, thus far the evidence is not on their side.
We in the sector talk a big game about “working ourselves out of business”. To the contrary, we have worked damn hard to stay in business while consuming dollars that are better spent by the poor themselves. As anti-interventionism grows, the social sector will have to more publicly reconcile its pro-social rhetoric with its own self-preservation.
I’ve long been fascinated by the simple concept of carbon offsets. A carbon offset is an arrangement where corporations “offset” their polluting by purchasing carbon credits that fund the development of renewable energy.
Without getting into the policy particulars of carbon offsets, the concept is straight forward and intuitive. Carbon offsets got me thinking about offsets more generally. Corporations, and individuals, do a lot more than just pollute the planet. Why isn’t there some sort of giving offset for everything?
While I was compiling my list of nonprofits to donate to for my year end giving I ran an experiment to see what an individual donor giving offset might look like.
Building a giving offset
For my personal finances I use the popular account aggregation service Mint. Since my finances are consolidated in Mint I figured it would be convenient to base my giving offset calculations using a transaction history exported from its website.
To automate the process of logging in and extracting transactions from Mint I used the excellent Python library mintapi. Mint conveniently categorizes all financial transactions, so I mapped each Mint category to a cause. For example, pet related transactions were assigned to animal related causes, education expenses were assigned to education causes, restaurant purchases were mapped to hunger, etc.
Based on the mapping of Mint categories to giving offsets, the system I built recommended I allocate my charitable giving for 2015 according to the following chart.
Given the current mapping and my personal spending habits, by far the largest offset recommendation is in the “Poverty and homelessness” category followed by “Food and hunger”. Readers of Full Contact Philanthropy might rightly wonder whether the current mapping is “fair”, or whether it instead reflects my own bias toward investing in poverty related issues. I don’t know the answer to this question, but my guess there’s a fair amount of bias (isn’t there in everything?).
Since Mint data extracts include one’s entire transaction history, not just the most recent year, I also ran the giving offset recommendations over the last few years as shown below.
As you can see from the above graph, there’s a fair amount of change in my giving offset recommendations from 2011 through today. Around 2013 suggested giving to animal causes grew considerably, reflecting vet bills one of my chihuahuas has been racking up the last few years. And yes, I took the giving offset’s advice and added animal related causes to my giving portfolio.
I’m not sure a giving offset is a good idea. I’m also not sure it’s a bad idea. I think in some ways I’ve only recently fully accepted and moved on from the closure of my last company Idealistics. This acceptance has resulted in a flurry of ideas and activity around those ideas (reflected partly in more writing on this site). The giving offset is one of those ideas.
However, I am committed to trying out various new ideas and putting those ideas out in public rather than keeping them to myself. I generally like the idea of charitable giving reflecting the life one leads and trying to offset selfish spending by re-investing in the world. I have some thoughts on how one might approach a more robust version of a giving offset, but I’m hardly married to the idea. What do you think?