Thinking Critically About ROI

by Brent Gloy and David Widmar

—

From time to time there are farm management ideas that get executed so poorly it makes the decision-making process worse. An all too familiar example of this is the Return on Investment (ROI) calculation. Maybe it’s just the two of us, but nearly every ROI calculation we see promoted, peddled, or printed makes us cringe.

Used correctly, ROI can be a powerful tool, but like most tools, the effectiveness is usually a function of the practitioner’s skill. With that in mind, we wanted to share five ideas to help you improve the effectiveness of using an ROI calculation.

Start with a purpose

ROI calculations are excellent for summarizing the returns of alternative investments. In many ways, ROI can be thought of as a scorecard for comparing two items that aren’t intuitively comparable. For example, Stock A was purchased in January for $425,000 and sold a year later for a gain of $39,000. Stock B was purchased at the same time for $10.50 and sold for a profit of $1.75. In this admittedly simplistic example, ROI creates a comparable measure of performance. Stock A had an ROI of 9.2%, while Stock B rang in a 16.7% return.

In agriculture, ROI calculations can be useful in many cases, such as determining the best seed hybrids, herbicides, fungicides, or livestock genetics. For each of these, there will be products with premium prices and promises of higher yields. At the other end of the spectrum, there will be options at a lower price point and with the potential for less productivity. A proper ROI measure can help sort out which is a better fit.

Finally, ROI calculations can help prioritize investments. Is it better to buy more farmland, invest in drainage tile, or build an irrigation system? These have very different investment profiles, and an ROI can help determine which investments should be pursued.

In short, before calculating or reviewing an ROI calculation, think critically about what you’re measuring and trying to size up. An ROI measure on its own has very little value. Oftentimes the greatest value of the ROI measurement comes from the process of pulling together the estimate.

Bimodal, asymmetric, and the illusion of average

Consider the classic fungicide problem. Let’s say, hypothetically, the ROI for a fungicide application is 14%. This might seem like a good investment, but every corn farmer intuitively knows it’s extremely unlikely to see a 14% ROI in any given growing season. Why? With fungicides, especially the formulations of 10 or 20 years ago, the yield response was usually much smaller or much larger than “average.” These returns were bimodal, meaning producers would either benefit a little (maybe even negatively) some years but have huge ROIs in other years.

Another example of bimodal returns might be betting a dollar while flipping a quarter. On average, you’re likely to break even, but in reality, you either win or lose a dollar each time.

The second situation is asymmetric returns. These are low probability outcomes that have huge outcomes. With these, the ROI calculation will look very small – or negative – but individuals will still sign up.

A great example of asymmetric investments made by a majority of people without much thought is home insurance [1]. Most pay their home insurance premiums and hope to never collect a payout. In fact, we hope for a negative ROI. However, it’s the large payout – and underlying home mortgage risk – that makes home insurance attractive, despite the ROI measurement.

In the case of bimodal and asymmetrical returns, an average return can be deceiving. To get around this, dig into the specific observations to get a feel for the range of outcomes. In the case of bimodal outcomes – such as fungicides – the frequencies and magnitudes are key. In the case of asymmetric returns – such as home insurance or winning the lottery – the underlying investment and payout details are key.

Measurement problems: Return over what?

Another oddity with ROI is getting specific about the costs and returns that are considered. Consider the following: you added a micro-nutrient to your fertilizer application that generated $50 per acre in additional revenue. Would you use the cost of the micro-nutrient for the ROI calculation of the total cost of production? In the first case – considering the marginal costs and benefits – you’ll likely generate a higher ROI, but one that overlooks all the costs you already sunk into raising the crop.

In the second case, a producer would measure the total ROI for the above scenario and compare it with the total ROI for a second scenario. The difference in the ROI between these two alternatives will likely be more subtle but can provide helpful context. While the marginal ROI might be high, make sure the gains are significant and effective for the big picture.

Keep in mind that – even with the same data and assumptions, two people can reasonably reach two different ROI calculations.

Measurement problems: We can’t capture everything

“Not everything that counts can be counted, and not everything that can be counted counts.” [2]

Most of the time, it’s easy to capture and quantify the big items, but what about the factors that are just difficult to quantify. Perhaps the hybrid yields the same and costs more but, based on personal anecdotal observations, dries down better at harvest. Or maybe hauling your cattle to the feedlot with your own equipment would be a higher ROI activity, but it would require a whole new set of management headaches.

The main idea here is that even if we square up the data and make good estimates and assumptions, there are factors that are simply too hard to quantify. One way to work around this is to think about key considerations ahead of time. Are there no-go parameters you can establish ahead of time? For example, perhaps you decide to avoid hybrids that are susceptible to cystic nematodes or certain diseases right off the bat. If we establish these boundaries ahead of time, they will be powerful in our decision-making process, rather than excuses used in hindsight.

What’s a good benchmark?

Without fail, every presentation we’ve given to a group of producers about ROI turns up the question, “What ROI should I expect?” This is a great question, but there are no easy answers.

In addition to all the issues already laid out, each producer will have their own risk and reward preferences. Additionally, we are all prone to accept a lower ROI with familiar investments. If you suspect the product is snake oil, you’ll need a high ROI threshold to get your attention.

Wrapping it up

A finance professor was on a flight and struck up a conversation with their seat neighbor. After learning the stranger made the financial decisions on which books a publishing house should fund, the professor inquired, “What is your ROI benchmark?” The seatmate laughed and said, “Professor, nobody calculates ROI, we focus on the payback period and total dollars earned in the first three years.”

While ROI is an important measure, it’s not an all-knowing, apex calculation that has clear insights. As a tool, the ROI can help decision-makers understand the costs and returns of a given investment. ROI measures are especially helpful when comparing alternatives. There are shortcomings, blind spots, and differences of opinion when it comes to calculating and interpreting ROI measures.

Perhaps the most valuable idea to keep in mind with any discussion around ROI comes from a former Purdue professor: “Test before you invest!” Whether you’re the decision-maker or the one trying to help said decision-maker, use the ROI framework for thinking through the situation. What are the costs, potential returns, alternatives, and unaccountable yet important factors? And as we mentioned before, the process of calculating the ROI – or at least thinking through the ROI assumptions critically – is perhaps the most valuable outcome.

—–

[1] Most insurance products have a negative ROI for the buyer. The exception would be insurances that are subsidized by the government, such as crop insurance.

[2] Often attributed to Albert Einstein, a quick web search suggests William Bruce Cameron first wrote this phrase in 1963. He lamented that “it would be nice if all of the data which sociologists require could be enumerated because then we could run them through IBM machines and draw charts as the economists do.”

 

—

Thank you for choosing Acuron® corn herbicide.  This AEI Premium article has been provided by Syngenta and the Find More Bushels program.

Acuron is a Restricted Use Pesticide.