# «Equity Valuation LinkedIn Corp José Miguel de Figueiredo Bettencourt Moreira da Silva Student Number: 152414022 Instructor: José Carlos Tudela ...»

4.2.5 Tax Benefits from Debt As discussed in the literature review the net value of tax benefits is debt times the tax rate applicable to the company.

The tax rate, as previously presented, will be the corporate tax rate of the United States of America which is set at 35% for LinkedIn.

LinkedIn’s current long-term debt, as stated in its balance sheet, comprises a bond issue of a total of $ 1.322,5 million. However, looking just at the company’s bonds doesn’t provide a full picture of LinkedIn’s interest expenses. A common practice of most companies is to camouflage their debt as operating leases, hiding these interests expenses as operating expenses and providing a much healthier perspective of its fundamentals.

Therefore, in order to fully evaluate LinkedIn’s debt its operating leases had to be converted to net debt. The first step was to calculate the company’s cost of debt, which is gathered by adding the company’s default spread to the risk-free rate. The default spread is obtained given the corporate debt rating of BB+, which puts its spread at 3.25%10. This value added to the risk-free rate leads to a cost of debt of 5.40%.

By applying this discount factor to operating lease information supplied by the company we reach a debt value of leases of $ 1.065,13 million, as can be seen on the following table.

With the interest expenses of each year, of both long term debt and operating leases we use the cost of debt to obtain the present value of these tax shields for each year and in perpetuity (in a method equal to DCF). This yields a total value of tax shields of $ 1.606,45 million.

4.2.6 Expected Bankruptcy Costs The final component in order to obtain LinkedIn’s levered firm value is the expected bankruptcy costs. As mentioned in the literature review, although the probability of bankruptcy can be obtained with some ease, bankruptcy costs are not so direct.

For probability of bankruptcy the corporate bond default rate was the chosen method.

Given a rating of BB+ this places LinkedIn at a default rate of 10% according to (Altman & Kishore, 1996).

In regards to costs of bankruptcy these can be divided in two groups: direct and indirect costs. Direct costs relate to lawyer and accounting fees and all expenses related to the administration of the bankruptcy. It is noteworthy to mention that according to (Warner,

1977) larger firms have considerably lower relative direct costs of bankruptcy. Whereas indirect costs refer to “lost profits that a firm can be expected to suffer due to significant bankruptcy potential”11. These costs are associated with higher costs of financing due to probability of default, loss of sales and in general all costs brought upon due to the bankruptcy scenario.

It is clear that calculating direct and indirect costs of bankruptcy is open to interpretation and varies sizably with industry and company size. For direct costs of bankruptcy, I refer to (Altman, 1984) where an empirical analysis of direct bankruptcy costs of retail and industrial companies yielded a range between 4% and 11%. Since relative direct costs have an inverse correlation with the size of the company, it is expectable that LinkedIn fall within the larger category and have around 5% direct costs.

The real challenge comes when assessing LinkedIn’s indirect cost of bankruptcy. Going by the same source we are pushed towards around a 10% indirect cost of bankruptcy, yet in this sense LinkedIn fits into a very different market segment from retail and industrial.

In fact, it is quite expected that LinkedIn have much higher indirect costs. A clear example Altman, E. I. (1984). A further empirical investigation of the bankruptcy cost question. Journal of

of this is the dot com crash and subsequent near total value loss of a large portion of internet companies at the time.

Having no real fixed assets on which to rely on and no production, an internet company in an unproven market segment, with mostly intangible assets and, fundamentally, a service provider is expected to lose most its value in a bankruptcy scenario.

However, given the lack of empirical evidence in this regard in the following chapter a sensitivity analysis will be shown where several costs of bankruptcy are tested to see the impact on the price of LinkedIn’s stock. For the purpose of this chapter an average indirect cost of bankruptcy of 55% was chosen.

In conclusion, with a probability of default of 10% and cost of bankruptcy of 60% the expected bankruptcy costs are given by the product of these two ratios and the unlevered value of the firm. This yields expected bankruptcy costs of $ 1.317,93 million.

4.2.7 Enterprise Value LinkedIn’s enterprise value is thusly obtained, as presented in the literature review, by adding the discounted cash flows (DCF), the terminal value, the interest tax shields (ITS) and subtracting the expected bankruptcy costs (EBC).

$26 000 $22 000 $18 000 $14 000 $10 000 $6 000 $2 000

As can be seen from the previous chart the value obtained is $ 22.254,08 million. From the breakdown of the enterprise value we can assess that the main component of this result LinkedIn Corp Equity Valuation is the terminal value, which accounts for 62% of the total. This puts most of LinkedIn’s value after its stabilization and dependent on its growth prospects and continued innovation in its sector.

4.2.8 Equity Fair Value Having calculated LinkedIn’s Enterprise Value, the next task is to reach its fair equity value (EFV), so that we can then compute its price per share.

The function is quite simple, just add cash and cash equivalents minus net debt (ND) to enterprise value (EV) to reach the equity fair value. As can be seen in the following

**equation:**

Given cash and cash equivalents (short-term investments were also included given their liquidity and the company’s own statements that for all intents and purposes they consider those investments cash) total value of $ 4.559,3 million and a debt level (long term debt plus operating leases converted to debt) of $ 2.390 million, this yields a EFV of $ 24.364,8 million. This value translates into a 4.8x terminal EBITDA multiple.

30 000,00 25 000,00 20 000,00 15 000,00 10 000,00 5 000,00

Chart 3 shows the breakdown in EFV. The main factor of note is that Net Debt is positive and therefore the EFV is actually higher than the EV. This comes from LinkedIn’s large cash reservoirs that are justified by the company as essential for its operations. Its low debt is related to low operating income (due to high investments) and therefore its optimal LinkedIn Corp Equity Valuation debt ratio is quite low (due to it being linked to its interest coverage ratio which for higher interest expenses requires higher operating income).

In order to obtain the EFV per share we divide this result by the total shares outstanding of around 130 million. This yields a result of $ 186,77 per share estimated for December 2016.

5 Sensitivity Analysis Given the uncertain nature behind the assumptions required for the valuation process it is standard process to run sensitivity analysis. Theses analysis consist in seeing the impact on the valuation of certain changes in relevant inputs. The aim is to give the investor a wider understanding of underlying risks faced by the recommendation and to be able to take them into account for his decision given his own risk profile.

Throughout this section several variables will be tested for the impact they have in the overall valuation in order to bring more robustness to this valuation and present the investor with possible outcome scenarios different from the base case presented in previous sections. On the one hand, these tests will serve to justify some assumptions made on the model. On the other hand, they also aim to study what can be considered possible and expectable variations, given the specific characteristics of certain variables.

This section will be concluded with analysis of the three scenarios mentioned in the revenues section. All sensitivity analysis before that section will be run on the Base Case.

5.1 Expected Bankruptcy Costs As explained in previous sections expected bankruptcy costs are difficult to estimate due to unpredictability in forecasting the impact of distress on a firm. The value chosen for bankruptcy costs was 60% and in this section we can see on table 7 the impact of variations on this factor.

Within 10% probability of default, which is the most robust of the two percentages given its empirical basis, variations of 20% up or down have only a $ 6 impact on the price. In the Appendix a full table of sensitivity analysis shows that significant impact on the price only comes when both variables change. In comparison to the base case of 60% and with 10% probability of default it can fluctuate from plus $ 8 dollars to minus $ 6, or plus or minus 5% of the base case value.

5.2 Cost of Equity and Growth Rate Arguably the factors that have the largest impact (with the smallest change) are the discount factors for the model. In the following table we will analyze the impact of a shift in both those factors simultaneously.

This scenario shows much more drastic fluctuations in the valuation then the previous case. The valuation shows more sensitivity to changes in cost of equity then stable growth.

Keeping growth at 2,8% and changing cost of equity yields variations between plus $ 32 and minus $ 23. It is also interesting to note that increases in valuation due to decrease in cost of equity is relatively larger than the negative changes from increased cost of equity.

The same can be said of the growth rate as can be seen from table 9.

Therefore, the biggest impact to LinkedIn’s value comes from the return demanded from its equity and not so much the impact on its steady state growth rate. A perception of riskier or less proven strategies from management or taking in much more debt than its current level could lead to investors requiring a higher return from LinkedIn and as can be seen from these tables that could lead to significant drops in the per share price of LinkedIn.

5.3 Normal Distribution Given the range of possible values and combinations from the previous sensitivity analysis it makes sense to run statistical analysis within the ranges believed more probable.

LinkedIn Corp Equity Valuation With a range’s mean and standard deviation several Monte Carlo simulations can be run.

A Monte Carlo simulation is (among other uses) a way to generate results within a probability distribution and therefore making it possible to take statistical conclusions.

In this particular case, and with the aforementioned data, various sets of 10000 results were generated. Such large samples were chosen in line with the Central Limit Theorem which states that given a sufficiently large pool of results its arithmetic mean will be approximately normally distributed. Therefore, allowing for our analysis regardless of the underlying distribution.

**5.3.1 Cost of Equity**

First range of values analyzed were a variation of 1% plus and minus the base cost of equity of 8%, given the same stable growth rate of 2,8%. Since this variable has the largest impact on the per share value of equity it is relevant to test an acceptable level of possible short term variation and its impact on the value of LinkedIn.

Mean StDev Loss Gain 188,78 21,66 99,5% 53,6% Table 12 - Summary results Cost of Equity distribution Table 10 shows some simple statistics related to this normal distribution, namely its mean and standard deviation. Assuming the 2,8% stable growth rate, the mean value expected with this variation actually puts the per share price of LinkedIn higher than the base case.

However, its high standard deviation is also a relevant factor to take into consideration, given that it represents nearly 12% of the base case prediction value.

Loss refers to the probability of, given this 1% fluctuation of the company’s cost of equity, that the per share value be lower than the closing price on December 1st. This result is relevant due to the fact that it points to close to 100% probability of losing value. Gain refers to the probability of the per share value be higher than the model prediction of $

186.77. The percentage of 54% for this statistic implies that the base case prediction actually stands on the bottom half of possible outcomes given the sensitivity analysis being run. However as stated previously this is a result of a decrease of 1% in the cost of equity having a relatively higher impact on the price than a similar change in the opposite direction.

5.3.2 Cost of Equity and Stable Growth A further test added a possible fluctuation in stable growth rate of 0,5% positive and negative added to the previous cost of equity variation. This test’s purpose was used to further analyze the impact on the value with a wider range of outcomes from the table and put into perspective a large possible result pool.

Mean StDev Loss Gain 190,00 23,68 99,0% 55,1% Table 13 - Summary results Cost of Equity and Stable Growth distribution Of note in this scenario is the increase in all statistics, in comparison to the previous section. Although this scenario included more outcomes, both positive and negative, the overall conclusion would be the same as the previous scenario. Clear current overpricing of LinkedIn’s stock.

5.4 Multiple Scenario Approach As stated in the revenue projection section, three scenarios were estimated for LinkedIn’s possible future revenue stream. The Base Case, which was followed throughout this dissertation.

The Bad Case Scenario analyzes how LinkedIn’s value would change if LinkedIn essentially failed as a SNS and converted completely to a PNS or job board. With little income coming from Marketing and no income from Learning.

Without synergy from LinkedIn’s SNS aspect Talent Solution wouldn’t increase as much, neither Sales Navigator.