Confronting the Question: How Much Capital Do I Need?

MAY 2021

A common question amongst LPs is: How much cash do I need to cover my capital calls in the next period? Many clients use the Takahashi-Alexander forecast model (already available in the Private i platform) or an internal Excel-based model.

A deterministic model such as Takahashi-Alexander’s, particularly with the parameters that Burgiss supplies, will generate the typical (or, more accurately, the average) capital calls for a future period. However, in order to include a confidence interval around the expected outcomes, you need a more sophisticated approach such as Cash Flows at Risk (CFaR). 

Cash Flows at Risk (CFaR), developed by the Burgiss Applied Research team, is a new simulation-based method for forecasting capital calls where the goal is to predict not only the typical capital calls, but also the range of possible capital calls for upcoming periods, combining Burgiss Manager Universe (BMU) and clients’ investment data. CFaR will help answer the question: “How much reserves are necessary to cover the capital calls in the upcoming period with a certain probability, such as 50%, 75% or 95%?”, or quite simply: “What kind of safety margin do I need?” 

Figure 1: Capital Call Probability Density Graph

Figure 1: Capital Call Probability Density Graph

 To further illustrate the value of looking beyond the mean, consider the stylized graph of a probability density function (Figure 1). The traditional model of analyzing forecasted capital calls returns the mean (the red line), which is calculated based on typical (or average) capital calls, while also not factoring in the benefits of a diversified portfolio.

The CFaR methodology factors in the idea that future capital calls are only partially predictable while also leveraging the depth of historical cash flow data in the BMU to build a histogram of possible capital calls. Using this histogram, you can estimate the 95th percentile (blue line), which we call the 95% CFaR. This number represents the capital reserve needed such that capital calls will be less than that amount with a probability of 95%. On the other hand, if you use the mean to estimate the capital reserve needed, that will result in approximately a 50% probability of not being able to service the capital calls, a risk that is unacceptable. 

The central idea of CFaR is to compute a specific percentile of possible capital calls for a private capital portfolio for an upcoming period, by simply replaying the capital calls of similar funds from a historical database, i.e. the Burgiss Manger Universe, and then to compute the percentile of the resulting replayed capital calls. The methodology, which is similar to what is termed a ‘historical methodology’ when measuring market risk, is based on the Burgiss Applied Research Paper Budgeting for Capital Calls: A VaR-Inspired Approach and the Cash Flows at Risk - Whitepaper. The important keywords behind the CFaR methodology are ‘replay’ and ‘similar’. 

We replay a fund’s past capital calls by treating them as a fraction of uncalled capital. For example, if a BMU fund with $100M of uncalled capital called $10M in a past quarter, then the fraction of uncalled capital would be 0.1. Therefore, we would replay that for an investment with $50M of uncalled capital as a $5M capital call, which is calculated as 0.1 * $50M. We consider a fund at some point in the past to be similar to a current investment if the fund belongs to the default peer group assigned to the investment and if at a point in the past, the age of the fund is similar to the current age of the investment. For example, if the investment is a 1 year old buyout fund, then all 1 year old buyout funds for each past quarter from the BMU will be considered similar.  

 Today, Portfolio Management users can apply the CFaR methodology to capital calls using the BurgissCalcCaR function in the Burgiss Excel Add-in. In order to compute a specific percentile of possible capital calls for a portfolio of private capital investments, the BurgissCalcCaR function, based on required and optional parameters, will perform the simulations described above and calculate the requested percentile in the Excel Add-in.

Figure 2: CFaR in the Burgiss Excel Add-in

Figure 2: CFaR in the Burgiss Excel Add-in

CFaR Webinar

To see the CFaR metholology in action, join us on Thursday, June 10th 2021 at 11:00am ET for a live webinar on Advanced Techniques: Analyzing Cash Flows at Risk in the Burgiss Excel Add-in. 

In this session designed for intermediate users of the Excel Add-in with a basic understanding of Burgiss peer group functionality within Portfolio Management, learn how to:

  • Apply the CFaR methodology to capital calls using the BurgissCalcCaR function available in the Burgiss Excel Add-in.

  • Leverage the Add-in’s formula to predict the typical capital calls as well as a range of possible capital calls for upcoming periods.

  • Understand the underlying calculation methodology powering the Cash Flows at Risk (CFaR) calculation.

Register here.