MMA of the metrics? CFROI vs. ROIC

With Carlo Eyzaguirre (and special thanks to Abhishek Saxena and Marc Goedhart )

Cash Flow Return on Investment (CFROI) and Return on Invested Capital (ROIC) are both metrics used to calculate returns for a given company / project. But which one should you use? As much as you might get that impression from fearless defenders of their method, It's really not the MMA of metrics. Both have their place, depending on what you are trying to achieve

• CFROI is a one-period performance metric that aims to estimate an IRR for existing assets. It uses a “pretend” investment into the assets as they are today (gross assets plus Working capital) as cash outflow, and projects current cash flows for a lifetime of the asset as returns (incl. inflation and releasing working capital at the end of life) as cash inflow. It relies on cash flows, not accounting earnings (but of course you need to estimate the cash flows, typically from accounting numbers, with many adjustments).
• ROIC is also a one-period metric that aims to estimate the underlying core operating performance of the business. You can think about it as an better way to estimate Return on Assets by adjusting for non-operating items. Ideally, it takes an adjusted post-tax operating profit (NOPAT) and divides it by the operating invested capital (essentially, net fixed assets plus working capital). It’s easy to calculate this as it is straight from the financial statements, easy to compare with peers outside-in, and can directly be compared to WACC[1]. On the other hand, it is relying on adjusted accounting earnings, often using a shortcut like reported EBITDA as a starting point.

?

So is one better than the other?

There are advantages to both, but they are very different:

• ROIC works much better for a business with many assets of different ages, and can much more easily be estimated and broken down into business segments – it’s basically simple to understand by anybody who runs a set of assets and has numbers without one-time charges.
• CFROI is much better to understand returns of single assets or sets of assets where profits are fairly constant based on the initial investment – for example, a (set of) power plant(s), pipeline (s), or windfarm(s) - where maintenance capital expenditure (capex) is much lower than depreciation, and profit growth is driven by gross asset addition.

The math behind it

To illustrate, consider first the asset in Exhibit 1, which is an illustrative model for a 5 year project where operational cash flow is related to gross invested capital. Depreciation is much larger than capex here - think about projects with big one-time investments followed by stable cashflows like infrastructure projects or hydroelectric power plants)

Exhibit 1: Stable cash flows following gross initial investment

The assets depreciate without capital replacement from maintenance or new investment, and the net book value goes down. This effect grossly inflates ROIC – all the way to infinity if the asset reaches a book value of zero and still produces income.

On the other hand, for this kind of asset, CFROI is constant over time since it uses cash flow and gross value. For a business of 5 assets started in 5 consecutive years (shown in the last column), the result is that CFROI matches the IRR of an individual project, while ROIC mathematically always ends up higher. CFROI is better to asses single assets or businesses of similar assets that have little maintenance capex needs so book value depreciates to zero while there are still earnings. Basically, when earnings are proportional to goss invested capital.

Now consider Exhibit 2. This is a company, which is typically (but not always) a set of many assets with many different service lives at the time of assessment. Future life is different by asset, often. Also, and most importantly, cash flows tend to relate to net investment capital as capital is redeployed: In the example, earnings go down as there is no additional capital invested and machines wear out or become obsolete. Think of retail supermarkets that have to refurbish shelf space, hotels that have to renovate rooms, or a manufacturing company with many plants and pieces of equipment. If there is no such capex, customers will not come and earnings go down.

Exhibit 2: A typical business where cash flow follows net capital

Here, ROIC is constant over time, but CFOI decreases over the asset lifetime as it relates the cash to the original investment. For a set of asset, the ROIC equals the IRR, but the CFROI is lower than the IRR.

Now what?

So which one should you use? Given the business examples shown above, we suggest using ROIC over CFROI when looking at the performance of businesses or business segments, and CFROI when comparing individual assets or group of assets where cash flow is driven by gross investments with little reinvestment needed to support the cash flow.

Why ROIC for typical companies as a whole, and not CFROI?

1. CFROI can underestimate returns for typical companies as shown above
2. ROIC can easily be broken down into segments. CFROI is hard to understand and to model for segments and sub-segments of businesses. The accounting adjustments and executive education is probably not worth the effort compared to a simple “operating income divided by net PP&E plus working capital” metric, which you also do pre-tax to keep it even simpler for BU management. (See below for some ideas to simplify CFROI)
3. Most perceived CFROI advantages are not quite as unbiased as they are often displayed and therefore not worth the time needed to implement. An IRR metric like CFROI still needs to estimate the economic life of the asset (i.e. the same estimation accountants do depreciation). Inflation adjustment is a great advantage if done correctly, but will probably not influence decision-making if it is low (as far as we can tell). And in high inflationary scenarios, this can hide some key assumptions as well.?

Why CFROI for individual assets, or sets of assets where gross capital drives cash flow?

The point of ROIC is to estimate a sustainable return on investments if investments continue. This mathematically just doesn’t work for a depreciating asset with constant cash flow, where CFROI shines. As shown above, the numbers get out of hand and don’t bear relation to the project economics as measured by IRR.

Simplifications used in practice can make CFROI easier to implement

• Using EBITBA/Gross assets as estimate for CFROI. In many projects with little additional working capital (since there is no growth) and little maintenance, this can be a great shortcut (see exhibit 1), but note that EBITDA/Gross Assets bears no meaning when compared to WACC, which is fine post investment decision
• Using "Economic Depreciation" instead of IRR. Without getting into details [2], this method uses the same idea but avoids the complicated IRR calculation. It does require an estimate for replacement cost of the asset(s), which again is relatively simple to do for a single asset like a windfarm but hard to do for a company with many assets and tough to break down from a corporation to segments.

?

What have you seen in practice?

Is your company using CFROI? If so, are you using it below the company level or in individual projects? Have you fallen into the trap of using ROIC for individual assets (we certainly see this a lot!)

?

?

[1] if done correctly, a company where ROIC is equal to WACC has an enterprise value that matches the capital exactly (as no value is earned over book value)

[2] This exhibit explains this alternative method for CFROI estimates