In light of buybacks being back in the news with the proposal to increase the tax, I thought I’d write about how companies should think about buybacks since many do it wrong.
First, quickly, buybacks shouldn’t be political. As you will see from my framework, they are a rational decision if done right. There is nothing inherently wrong about them and they don’t deserve government scrutiny.
Also, these principles can be applied to any company, not just insurance. Just substitute ROIC for ROE and EBITDA multiple for book value multiple and you’re good to go.
Rule #1: You can’t just look at valuation
How many times have I heard “The stock is at book. Why aren’t you buying it back?” or “The stock is down 30%. Why aren’t you buying it?”.
That’s irrelevant. What if you’re at book, but your ROE is 6% with no prospect for improving? Pay dividends. A buyback isn’t creating value as the company deserves to trade below book.
Rule #2: It doesn’t matter if it’s accretive
Buying back stock to drive EPS is a dumb reason to buy back stock. Why? Because it uses the unrealistic hurdle rate of cash.
In other words, taking cash, which is earning a low single digit return, and using it to buy back stock will almost always drive up earnings (unless you have a very high P/E).
That same faulty logic also leads companies to justify cash-funded M&A. Accretion of M&A, buybacks, or anything else needs to be compared to a fully loaded cost of capital which leads to our next rule.
Rule #3: You need to consider the alternatives
A company has effectively four options to deploy capital: invest in organic growth, acquire something, pay a dividend, or repurchase stock.
When you choose to repurchase, you are effectively stating those other three options aren’t as good. That may well be true, but it requires a proper framework to evaluate.
Organic Growth
Most companies #1 reason to not buyback stock is “we have better organic growth opportunities”. This may well be true, but it is also true that most companies overestimate the profitability (and underestimate the risk) of their organic growth.
To use a simple example, if a stock is trading at book value and has a steady 12% ROE, then repurchase “earns” a 12% return. So, if there is an organic opportunity at 15%, by all means, the company should prioritize it.
However, often the organic growth is at 10%, but the CEO would prefer to grow or pursue some preferred project and so they take the lower return option. This is clearly value destructive.
The gray area is when organic growth is say 13% but has execution risk. While I don’t fully agree that buybacks are “risk free”, they are certainly low risk.
Likely the best way to account for this is to run scenarios of outcomes for the organic project and see what the probability is it does better or worse than buyback under different outcomes.
If say 2/3 of the time, the IRR is > buyback, then proceed with the project.
M&A
One of the more common reasons companies give for avoiding buyback is they prefer to use excess capital for M&A. This is very reasonable – as long as the M&A actually earns a decent return.
Again, EPS accretion is the wrong way to look at buyback. Anything can be made accretive with aggressive enough financing.
The best way to judge accretion is whether earnings will be higher (say in five years) if the company goes forward with the deal or, instead, deployed a similar amount of capital into buyback.
There are some challenges with this approach on larger deals that require equity (since if you do too much buyback, your equity shrinks to the point where your debt/equity explodes and your credit gets downgraded), but for small and mid-sized deals comparing M&A to buyback is the proper framework.
Dividend
If organic and M&A aren’t good options, then the remaining debate is dividend vs. buyback. This is very simple. If the valuation is “high”, you do a dividend. If it’s “low”, you do buyback.
How does one evaluate that? Keep reading.
Rule #4: Have an excess capital strategy and stick to it
If buyback is the best option, there should be a consistent decision process around how big a program to authorize relative to the amount of excess capital available.
Buybacks are no longer low risk if they increase the risk of financial distress. We have seen insurers lever up to buy back stock, then a cat hits or they have large investment losses and they are issuing equity far lower.
The best way to avoid this temptation is to create rules around excess capital buffers to ensure safety. Many companies do this, though sometimes they are not dynamic enough (i.e. they don’t contemplate shocks to the required capital).
Rule #5: Have a valuation strategy and stick to it
I’ll cover this more below too, but if you understand your IRR on organic growth and M&A, then you should be able to calculate what valuation ranges you would buy back stock.
This range should be communicated to investors and serve as an upper and lower bound for repurchase activity. For example, “we will not repurchase above 1.5X BV, we will repurchase aggressively below book, and will vary our activity in between that range”.
One of the biggest mistakes managements make is continuing to buy back (or even increasing activity) as valuations rise. Like clockwork, companies buy the peak on their stock and then are too scared to buy the trough.
This is usually because they are worried about signaling their stock is expensive if they stop buybacks. Conveying a rules based framework for buyback activity gets management out of this trap.
Rule #6: Employee option dilution is irrelevant
If employee options increase share count, then that is a real cost of issuing those options. If the stock is expensive or there are better growth opportunities, it is foolish to have a buyback just to sterilize the option dilution. It is an inefficient use of capital. Better to let the share count grow in that case.
Building A Framework
Now that we have our rules, we can construct a clear buyback philosophy.
Step 1: Evaluate the IRR of organic and M&A opportunities
If there aren’t good opportunities to deploy, then the discussion is pretty simple. We move to dividend or buyback depending on valuation.
If there are good opportunities, then we need to move to step 2.
Step 2: Compare organic/M&A IRR to buyback
As discussed above, growth opportunities should be compared relative to buyback, but how do we do this?
My preferred technique was to model out the financials first with the growth opportunities and no buyback and then again without them and using all of that capital for buyback.
You then see which one produces the highest future (5-10 years out) earnings and book value. That’s your winner. It’s really that simple. Everything is about per share returns.
While you can’t shrink your way to greatness, $ of earnings isn’t the goal. The goal is $/share earnings. A CEO is a fiduciary for the shareholders, not the senior VP who wants to manage a bigger footprint.
Step 3: Evaluate excess capital
If there is enough excess capital to pursue all acceptable growth opportunities, there may still be room for buyback afterwards. If not, then growth capital takes priority over the buyback.
Step 4: Consider valuation
This is where we decide if capital return should come as a dividend or buyback. There are a few ways to approach valuation.
If you have a good intrinsic value model, you can have a rule to buy when you’re at a discount to that. This is the approach Berkshire and others use, but I am skeptical most companies can do this well.
An alternative is to flip the M&A approach from above. So you model a reasonably valued theoretical acquisition of the same size as the contemplated buyback.
So if you are looking at a $500M buyback, model a $500M fake deal. Then, model the $500M buyback and compare the long term difference in earnings and book. If the buyback is better, you go forward. If it’s worse, you either sit on the capital and wait for a deal or pay a dividend.
In case it’s not clear, this approach will lead you to do more buyback when your stock is cheaper (and/or your ROE is higher) because you will be able to buy more shares for the same dollars, so the odds of beating M&A are higher.
Finally, you can consider a historical valuation approach where you buy at the bottom of the range and pause at the top. This is tricky though as it assumes your forward prospects are as good as those in the past. It also is a declaration that the market’s valuation is wrong, which may be a blind spot to a problem the market sees.
Step 5: Translate the approach into one investors can easily grasp
Regardless of which approach you use, you need to then have a simple way of communicating it to shareholders.
If you have an intrinsic value model, it likely corresponds to a P/B valuation. If you model vs. M&A, it will look worse at higher valuations and better at lower ones.
The key is to communicate these ranges to shareholders as proxies for your internal approach. If investors understand at what prices you will be more and less aggressive around buyback, you will not get stuck in the signaling trap where you feel pressured to buy the stock when it’s expensive.
They will understand that at certain valuations you will transition from buybacks to dividends. Note, there are a number of highly valued insurers that do special dividends instead of repurchase. The “signal” that their stock is expensive hasn’t hurt them.
Dumbed Down Version
If all that is too long and you need a simpler way to remember, follow these steps:
- Are there high return growth opportunities better than buyback: Yes or No?
- If no, how much excess capital can we comfortably return?
- Should that return be a repurchase or a dividend based on current valuation?
It’s really that simple, but it’s surprising how few public companies have a discipline like this.
More importantly, it creates guardrails to prevent value destruction through expensive buybacks and makes it easier to justify increasing the buyback when the stock is cheap.