Control Premiums

In valuation, there are times that some practitioners believe it is appropriate to apply the so-called “control premium” to their valuation process. While I have mixed thoughts on the use of this control premium, I would like to explain the concept along with my thoughts on when and when not to apply it. I will also shed some light as to the reasons for my mixed thoughts on the use of control premiums in practice.

To begin this discussion, I’d like to talk about what a control premium theoretically is. To help better explain this think about yourself as an investor in Microsoft; as an individual shareholder it is likely that you have little to no influence over the direction of the company. This is because of the sheer size of the firm and your likely minority position. In the event that you had the ability to control the firm you would theoretically be able to make certain decisions that are in your best interest, as an equity owner, and thus be willing to pay a premium. While this is a very simplified explanation, I would like to take a closer look at why one might be willing to pay this premium.

Take for example a firm such as Jos A Bank, a men’s clothing retailer who has a balance sheet loaded with cash and virtually no debt. Corporate Finance 101 teaches us that the cost of debt is almost always cheaper than the cost of equity. The second part of this lesson is that adding debt to the capital structure, to the extent that the additional risk of bankruptcy is offset by the cheaper capital and tax benefits associated with debt, will lower the cost of capital for the firm. Taking this lesson and applying it to the Jos A Bank situation, a controlling owner could increase the use of leverage within the firm and therefore lower the cost of capital and increase the number of profitable projects the firm can pursue or the net present value of their current projects. In essence, having this controlling interest allows for an increase in the value of the firm and thus the purchaser would theoretically be willing to pay a premium from the current minority stake price.

In valuation, there are three primary models used in practice to determine your final valuation: discounted cash flow, comparable companies and precedent transactions. While I do not want to get into the details in this post, I would like to discuss each of the three and why or why not you would potentially apply a control premium.

When looking at a discounted cash flow valuation model, one is forecasting future economic benefits to an ownership stake and then discounting these benefits to today’s dollars using an appropriate rate that reflects the risk associated with the cash flows. Now if one believed that they would be able to achieve more economic benefit from the firm by gaining a controlling interest, and intended to gain this control, they would be able to forecast the additional economic benefit into the model. This would then be discounted and the final value would already factor in the added value for control, thus a control premium would not be appropriate for the discounted cash flow model.

Comparable companies model takes a set of firms which are similar to the firm being valued and analyses key multiples. For example, if firm A is comparable to firm B and firm A is currently trading for 7.5 times EBITDA, then it might be safe to say that firm B is also worth 7.5 times EBITDA. There are many details to this valuation process that I will not cover in this post, but the key take away is that you are comparing the current trading price of comparable firms. Back to the original discussion about Microsoft, most shareholders are minority shareholders. Thus, the currently trading price would generally represent a minority interest and thus one who intended to purchase a controlling interest would then apply the control premium to the comparable companies model.

Lastly, precedent transactions takes a look at historical deals done in the open markets in which a buyer purchases a firm similar to the firm you are valuing. For example, firm C was purchased for 8.0 times EBIT, and firm D is similar to firm C. Thus, firm D might be worth 8.0 times EBIT to a purchaser. Now the key here is that this is based on a purchase price; implying that control is achieved. Therefore, the price paid would already have a control premium factored in and thus a control premium should not be added to the precedent transactions model.

To summarize thus far, it would theoretically be appropriate to apply a control premium to the comparable companies model only, but not to the discounted cash flow or present transactions model. Now enters the question as to what premium to add if you do intend to apply this. In practice many will rely on the data that is provided by Mergerstat for control premium. This can range from about 20-30% on average and is based on the premiums paid on historic transactions over a selected time horizon. This is where I find myself torn on the control premium.

I’d like to reflect on what a control premium is; recall that is the additional price one would pay for control due to the added value that comes from control of the firm. Also recall that if we used intrinsic valuation methods such as a discounted cash flow, we would factor this added value into the cash flows. One could value the company as it stands using the discounted cash flow model then value the same firm again but this time applying the added future income from the ability to control the firm. The differences in these two valuations would theoretically be the value of control, or the control premium. Mergerstat data on the other hand is simply historical premiums paid, but fully independent of the current firm and potential control benefits. I believe that it would be more advantageous to use the intrinsic valuation method to determine the value of control rather than simply applying the historical average. Also, remember that Mergerstat data is a statistic that has its own standard error which can be quite large and make confidence intervals wide.

This is not to say that the other side of the argument is not strong as well. First, conducting two discounted cash flow valuations with any level of accuracy will likely be very time consuming. Secondly, these models are based completely on forecasted data, which has a level of uncertainty associated with it. Lastly, it is very common in practice to apply the control premium based on Mergerstat data, thus it is hard to ignore widely-accepted practices.

In conclusion, I have used the terms “theoretical” often in this post to stress that I do not have a clear view on the proper way to handle this in practice. However, I do believe that this helps explain the basis for both sides of the argument and allows one to make their own decision on the use of control premiums in valuation.