Making TSR analysis more insightful

By Werner Rehm and Matheus de Oliveira Ramos

We, along with investors, executives, and analysts, use total shareholder return (TSR) as a metric to evaluate the performance of a company over a longer period of time, but the most common way to analyze the drivers of TSR?can fall short. ?

Usually practitioners divide TSR into two components: dividend yield in the period and change in share price. The latter is then represented as a change in earnings per share (EPS) and multiple.?

However, this method ignores some core principles of value creation:??

1. Dividends don’t create value for shareholders – rather, they are a way to pay back the value in cash. (We have written about this before: Theory: Value from share repurchases and dividends and Empiricism: Value from share repurchases and buybacks.)?
2. Simply looking at EPS growth and change in multiple doesn’t accommodate the fact that not all growth creates the same value. In particular, growth requires investment, sometimes at low net present value. (You can read more about that topic here: Organic or inorganic growth: Which generates more value?)?
3. EPS growth is a function of revenue growth and margin expansion – and also of nonoperating items such as interest expense. Thus, it can be far from the underlying operating performance. In general, it’s fair to say that EPS, as a standalone measure, doesn’t work to measure value creation.

Disaggregating TSR into real financial drivers

The better approach to analyzing TSR is to disaggregate it into its three underlying drivers – the impact generated by growth, profitability, and investors’ expectations, as measured by the multiple change (Exhibit 1). This can be done using a mathematical approach:

1. First, it’s possible to measure TSR in absolute terms using the change in market capitalization plus total dividends paid. Note that you will need to adjust for buybacks when doing this, though, if they are significant.
2. The change in market cap can be disaggregated into net income and the change in the multiple, with net income growth being separable into the revenue growth and change in margin for the period. This is very similar to the traditional method.
3. Now note that dividends are net income minus investment into equity. If you divide this equation by the market cap, you get two interesting components: (a) Net income divided by market cap, which is the inverse of the price-to-earnings multiple (P/E). This is a “zero growth yield,” as it represents the company’s TSR if it had no earnings growth and investors had no change in expectations. [1] (b) Change in equity divided by market cap– this is the reinvestment into the business. In your models, you will also need to adjust for other (non-cash) items that you want to take out that might influence equity (e.g., using the statement of comprehensive income).
4. If you want to go one level deeper, you can disaggregate the reinvestment into the required reinvested amount (for example, as estimated by an equity/revenue ratio for the organic growth) and the change in ROE/ROIC.

Exhibit 1.

Leverage

So far, we have discussed doing this on an equity level. However, two companies can have different TSR even though they have equal underlying value creation because of differences in their debt-to-equity ratio.

To do the analysis on an entity basis, you can use the fact that the TSR of a levered company is related to the TSR of an unlevered company via the market-based debt/equity ratio, and express TSR in terms of an enterprise value multiple, invested capital, and a leverage component. [2] We are using EV/NOPLAT, change in invested capital, noplat margin, and revenue growth, but you could use another set, as long as you are consistent between profit, multiple, and capital metrics?

Let’s do it

This real-life example looks at two companies?with a high TSR (Exhibit 2). It becomes clear that the drivers for the high TSR are very different. Company A’s performance is driven mostly by the change in investor expectations. The margin has compressed, and despite the lower growth, the resulting performance is impressive. This company’s management has to make sure it understands investor expectations and can successfully perform against them. The new multiple implies a margin expansion back to historical levels and potentially higher growth.

On the other hand, Company B has a positive TSR from the operating performance alone, driven by less margin compression at double-digit growth. The multiple expanded but not quite as much.

Exhibit 2.

Both companies in this example are software-based and asset-light, meaning the required investment is not significant, and their exposure to the travel industry during COVID-19 may explain the margin compression. Additionally, since neither company had any relevant M&A during the period, the increased goodwill (separated out from the investment) is close to zero for this example.

In future posts, we will look at more examples using this methodology.

[1] Think of a company with a $100 market cap, $10 net income, no growth, no change in margins or multiple. This company produces the net income in excess cash, as capex and deprecation cancel out and no other account changes. This cash can be paid out as dividends and is exactly net income/market cap if expressed as yield. This will happen every year.

[2] This sentence does a lot of lifting, but you can do it!