Insights Blog

Spending Policy | Old Technology Still Dominates

Written by Timothy T. Yates, Jr. | Mar 3, 2023 2:06:10 PM

Every year I eagerly await the release of the NACUBO Study of Endowments (NTSE) and, when I finally get it, I read it cover-to-cover. Despite reading all 195 pages, the two most frequent questions I get are: What was the average one-year return and how did we do versus that average return?

These questions often inspire me to rant, sorry, write a blog, about the inverse correlation between interest in a one-year return and relevance of a one-year return. High level of focus and interest; low level of relevance. One such blog was published during a particularly good year, but the fundamental points remain the same in relatively challenging years, such as FY2022. While reading this year’s study, though, my focus shifted from the inevitable questions about one-year returns to a much more troubling trend that I see in the data. What is most troubling about this trend is that it hasn’t changed much in the last two decades. It is a trend that has remained remarkably, perhaps stubbornly, unchanged over the years. So, instead of suggesting that you resist the temptation to celebrate a good one-year number, or bemoan a bad one, this year I ask that you focus on something else: your spending policy. Why? Because so many of you continue to use an antiquated “technology” that was developed half a century ago.  

I did my first work on spending policy over 20 years ago under to the tutelage of the late Commonfund President, Verne Sedlacek. Verne had come to us from Harvard Management Company and was immediately a towering presence at our firm. With that presence came a deep understanding of but also curiosity about the financial management of nonprofits. That curiosity led him to research and ultimately write a white paper for which he tasked me to do the analysis. As a result, I built the spreadsheets that underpinned his analysis and, in the aftermath of the recession in the early 2000s, we published a seminal paper entitled “Why Do We Feel So Poor?” It was a challenging period for nonprofits after the boom of the late 1990s and many found their spending payout falling significantly and impairing their operations and/or missions. 

In that paper, we discussed intergenerational equity, but also the volatility of spending and the impact of that volatility on nonprofits. We proposed considering other spending methodologies that might change the volatility characteristics of spending dollars. At that time, according to the NACUBO study 75 percent of colleges and universities utilized some form of a rolling average to calculate their spending, with the majority using a three-year or 12-quarter average. Our premise was that although three years would smooth spending to a degree, it still exposed the institution to all the volatility of the capital markets. This dynamic played out dramatically in 2003-2005 as negative market values rolled into the formula, resulting in forced reductions in spending. 

Since that time, Commonfund has published many papers, and given many presentations on the concept of decoupling or delinking spending for institutional need from the volatility of the capital markets and yet today 75 percent of the NTSE FY2022 respondents still reported spending a percentage of a moving average of the endowment’s market value. Imagine for a moment that you surveyed 700 of your friends and almost 525 of them told you they were going to a physical store to buy Taylor Swift’s latest 8-track.  

So why do so many institutions still use an old technology? The answer we get most often is that it’s easy to explain to donors, easy to use and it’s not too complicated. We all know that person who still uses a flip phone, but here we’re talking about using a rotary phone with a (gasp!) cord attached to it. Or trying to watch TV on a black and white box with a dial and only three stations. Not to belabor the point but many of the largest endowments utilize different spending policies that create less volatile payout streams and as far as I can tell, based on the amount of funds they raise every year, have no problem explaining or implementing their spending policies.

In theory, we want to take as much risk as we can in our endowment portfolios to generate the returns we need to fund our missions. Or at least we want to take as much risk as we think we can get compensated for taking. And our endowments, as long-term perpetual pools of assets, should be able to withstand short-term volatility, as one measure of “risk.” However, we certainly don’t want to transfer that risk to our institutions. The spending policy is that transfer mechanism. It transfers risk from the endowment to the institution and hence it is critical to understand and structure it in a way to minimize that volatility. 

Most recently, my colleagues, Anthony Peretore and Rachel Clivaz authored another paper on spending policy in which we offered several conclusions, including the idea that spending policy could be used as a risk mitigator in constructing an asset allocation. Spoiler alert: in most cases, hybrid, weighted average, or banded-inflation methods can reduced spending volatility – see their paper for the full analysis.

As we all look to the next 10-to-20 years and try to generate the types of returns we need to support our missions, spending policy will be more important than ever. So, as you read your copy of this year’s NTSE ask yourself why three-quarters of the respondents still use such an antiquated spending policy. If you are one of the nearly 500 who do, and the reason you have not embarked on an evaluation of an alternative spending policy is indeed because it’s “too hard to explain,” please call us! We would love to engage with you, your committees, your boards, your development team, or whomever at your institution wants to listen.