Nowcasting Methodology: Organic Growth

Ajit Agrawal, Samarth Nagaraj*, Neeraj Sudhakar

AKAnomics Nowcasting process relies on global industry data being able to explain changes in each company’s business revenues, especially for companies in cyclical industries (e.g., Industrials, Consumer). In order to build the relationship between changes in global industries and changes in company revenues, AKAnomics focuses on “organic” revenues of companies rather than “total” revenues. Most companies do not provide organic growth data, and when they do, their definitions of organic growth are not always consistent with what we need. Hence, building historical organic growth series on each company requires analyzing historical filings, especially as it relates to non-organic portion of the company growth (e.g., M&A). The question that clients often ask is if we could have gotten away with simply using the revenue growth data series (which is easily available).

This concept of company Nowcasting is new to the industry, and few hedge funds, if any, have enough experience in this field. The few funds we know of that have attempted a solution similar to that offered by AKAnomics Nowcasting, have done so by looking at total growth, but acknowledge the shortcomings of their process.

We believe that having clean historical organic growth data is critical to company Nowcasting. To explain our rationale, we analyze the historical data for the ~120 companies we track, and show that the non-organic growth (e.g., acquisitions & divestitures, currency) is a material component of revenue growth. Since acquisitions and divestitures (M&A) cannot be modeled through global industry data (its correlation is 0% with global industrial production data), this component can be thought of as a fixed error to a Nowcasting modeling process if not modeled separately.

The chart below shows the contribution that average M&A has to the average revenue growth across all the companies we track. It shows that M&A contribution varies anywhere from 20-40% over time (we focus on the changes in growth rather than growth since it has better statistical properties), and hence is a significant driver of revenue growth, and not separating out from the modeling process is akin to a “garbage-in garbage-out” approach.

*Samarth Nagaraj is a Summer Intern at AKAnomics Inc, and we thank him for his contributions. Samarth is an undergrad at George Washington University