U.S. Bank of Washington & Redhook Ale Brewery Co.



The Fuqua School of Business at Duke University

FUQ-04-2010

Rev. April 30, 2010

Autopista Central, S.A.

The Valuation of a Toll Road Project in Chile

TEACHING NOTE

Synopsis

This case is set in mid-November 2007, with Manuel Flores, senior vice-president of business development for the Spanish construction and services company ACS, evaluating a proposal by the Spanish toll-road operator Abertis and the private equity arm of the Spanish bank Santander to buy ACS’ 48% equity stake in the Chilean toll road Autopista Central. The would-be buyers are operating on a tight schedule and therefore want a proposed sale price from ACS by the end of the year. To provide a price that is reasonable to both sides, Flores will have to determine an appropriate discount rate to value Autopista Central’s future cash flows as well as consider the strategic goals of ACS while factoring in the prevailing macroeconomic environment of late 2007.

Two factors make Flores’ valuation of Autopista Central difficult. First, Chile, although one of the wealthiest countries in Latin America on a GDP per-capita level, is still considered an emerging economy which is characterized by a number of market inefficiencies. These inefficiencies violate many of the assumptions of the Capital Asset Pricing Model (CAPM) which make the utilization of traditional CAPM based methods inappropriate for effects of calculating a project’s cost of capital. If he could not use the CAPM, what model should Flores use?

Second, besides considering the intrinsic value Autopista Central’s cash flows, Flores must think about the timing of the proposed sale. Specifically, the run up to the time of the case was characterized by a long period of historically low interest rates and record levels of liquidity leading to a global real-estate bubble. As a result, going forward some speculated that world prices would fall sharply while others argued that new real-estate price levels were the “new normal.” How would the state of the macro environment affect the price that Flores asked for?

This note was prepared by David Keller, Belisario Galarcep, Massimo Paone, and Javier Vilardell under the supervision of Professor Campbell R. Harvey for the sole purpose of aiding classroom instructors in the use of Autopista Central, S.A., The Valuation of a Toll Road Project in Chile, FUQ-04-2010. It provides analysis and questions that are intended to present alternative approaches to deepening students’ comprehension of business issues and energizing classroom discussion.

Copyright © 2010 - all rights reserved. The Fuqua School of Business, Duke University, 1 Towerview Drive, Durham, NC 27705, USA. .

Pedagogical Objectives

This case is part of a module on calculating the cost of capital in an emerging market in the elective curriculum course entitled Advanced Topics in Corporate Finance. It was designed for MBA students and executives with either a financial or strategic interest in emerging-market investments. For this reason, it is appropriate for business, public policy, government, strategy, and international finance courses. The case has three pedagogical objectives:

1. It discusses the valuation of a long-term project in a developing country environment and highlights many of the common emerging market-risk factors that a manager must consider when calculating a project’s cost of capital in such a setting.

2. It presents the Damodaran Modified World CAPM and Erb-Harvey-Viskanta Cost of Capital (EHV) models as an alternative to the traditional CAPM given that this method relies on assumptions that are largely absent in emerging markets and therefore have limited practical value in an emerging market context.

3. It considers some of the qualitative factors that a selling party should consider when considering the sale of an asset such as strategic goals and market conditions.

Substantive Analysis

Class discussion falls into three main topics: (1) an initial discussion of the history of the project including a description of the current owners and proposed owners, (2) how the project’s risks and other inputs should be translated into the Damodaran Modified World CAPM and the EHV cost of capital calculator (Note: The accompanying Microsoft Excel spreadsheet file is an integral part of this case and necessary to complete this case’s cost of capital calculation) to produce a range of appropriate discount rates for the project and (3) the derivation of an appropriate range of values for ACS’ stake in the project. The class ends with a final discussion about the price that ACS should ask for its stake in Autopista Central.

Project History and Background

To begin the discussion, the instructor should ask members of the class to give a brief synopsis of the Autopista Central project. The audience should be able to recall some very basic facts about the context of the project. Some of the particulars that are of importance include the state of the Chilean economy in 2007 (over a decade of democratic rule, history of free-market reforms, reputable central bank, moderate but steady inflation), the mechanics of the Autopista Central toll road concession (concession of 30 years, construction fully completed by April 2007), the location of the project (Santiago which accounts for roughly 43% of Chile’s GDP and where around 40% of the total Chilean population lives), and the concept of the project (first free flowing toll system in Chile which connects to other major Chilean highways).

The instructor should discuss with the class the ownership structure of Autopista Central by referring to Exhibits 4a and 4b. Upon doing so we confirm that the company is effectively owned 48% by ACS and 48% by Skanska with the remaining 4% of shares being held by two Chilean construction companies.

Who are the common equity owners of Autopista Central and what are some of their interests and motivations? Both Skanska and ACS are very similar in the sense that both are large multinationals with a broad experience in a number of emerging markets like Chile. Both companies have a little over 50% of their voting shares floating in public markets although ACS has only four shareholders of non-public shares that control 48.5% of the company. In addition, most of the non-public shares of Skanska are held by institutional investors. Both companies are better known for their construction capabilities than for their post-construction management of projects.

The class should be able to describe the details of the capital structure of the project. For example, total invested capital is $815 million which is composed of $87 million in common equity + $603 million in debt + $125 million in subordinated debt. What are the characteristics of the bonds that were issued to finance the project? If the students translate all of Exhibit 11 to USD, they will get the following:

This table will be useful later on when determining the project’s weight of debt in the capital structure which will be needed to value the equity of Autopista Central.

Finally, the instructor should briefly mention something of the interests of the proposed buyers of ACS’ equity stake. To begin, why does Abertis want a piece of Autopista Central? Besides being a company whose core operating strength is the operation of public infrastructure projects, the firm is also looking to add complementary assets to its Chile infrastructure portfolio which includes two large toll roads, parking lots, a logistics center, and the Santiago airport. Why is Santander so keen on investing in the toll road project? At this point in 2007 Santander Private Equity is looking to build up a pipeline of investments that it can later pass on to the proposed $2.1 billion Santander Infraestructuras II (S2) fund. To do so the bank is eager to put up its own capital immediately. Also, although not implicitly stated, the fact that all three firms – ACS, Abertis, and Santander – are Spanish makes doing business culturally, linguistically, and perhaps fiscally favorable.

Autopista Central’s Cost of Capital

To estimate the value of the project we recommend using the discounted cash flow (DCF) method. In addition, we will estimate the internal rate of return (IRR) to incorporate in the analysis. In the case of the DCF, we will discount Autopista Central’s cash flows by the project’s weighted average cost of capital (WACC) which incorporates the total costs to the firm for both debt and equity capital. Consequently, the cash flow projections to be discounted will be the total cash available to both debt and equity holders.

The Discount Rate

The effective discount rate or cost of capital of the project is composed of the cost of equity and the cost of debt (adjusted by the tax shields that the firm gets from the interest payments on debt).

WACC = re x E/(D+E) + rd x (1-tc)x D/(D+E)

Where:

re: Cost of equity

rd: Cost of debt

E: Firm equity value

D: debt value

tc : corporate tax rate

In this case we are assuming that the equity value and debt value in the projected financial statements reflect market values.

1. Calculating the Cost of Capital

To obtain the cost of equity of the project the firm has many CAPM-based alternatives to choose from. But, considering that in emerging markets risk can appear in different ways and can affect a project in different ways as well, the Erb-Harvey-Viskanta Country Risk Rating method is the recommended model to use to best incorporate an array of risks into Autopista Central’s cost of capital calculation.

Parameters involved in the cost of equity calculation using the Erb-Harvey-Viskanta Country Risk Rating method were the following:

a) U.S. Risk Free Rate:

The average US Treasury Long Term Rates from daily rates are the following.

b) U.S. Equity Risk Premium

Based on the CFO Survey about Equity Risk premium, the 2007 reports shows that the premium over the 10-year Treasury Bond is 4.58%.

c) Institutional Investor Sovereign Credit Ratings

According to Institutional Investor, the United States and Chile had the following Credit Ratings in 2007.

[pic]

d) Beta (β)

The beta of the project measures the risk of the project relative to the market risk. In the case of Autopista Central, Exhibit 15 provides a list of the betas of comparable firms. The median beta from this list is 0.85 which we recommend adjusting according to the Morgan Stanley beta adjustment method which incorporates mean reversion to the beta of the market (i.e., 1). The Morgan Stanley adjusting formula is the following:

β = [βEstimate x (2/3)] + [ 1 x (1/3)]

Where:

β: Beta of project

βEstimate: Estimated beta of project from comparables or other method (e.g., OLS regression)

Upon doing so, we arrive at a project beta of 0.90. Nevertheless, based on the analysis of the factors that affect the project, we believe that the beta should be even closer to that of the market (i.e., closer to one). As a result, we used a beta range between 0.90 and 1.10 in both the Damodaran Modified World CAPM and the EHV cost of capital calculator which, together with the other inputs, yields different costs of equity.

The factors that we analyzed for the beta were the following.

i. Business: in terms of business risk, it has a medium level because of the following factors.

• Stage Risk: The toll road is in its post-completion stage, the toll road has been completely operational since 2006.

• Revenue Risk: The implementation and use of an electronic payment system has been done successfully. Thus, Revenue Risk is expected to be low.

• Traffic Risk:

o Santiago represents 32% of the overall Chilean population.

o The GDP of the zone encompasses approximately 40% of the total Chilean GDP.

o Users of the North and South corridor represent 35% of all the city vehicles.

o At least 65% of the vehicles from the Metropolitan area pass through the North and South corridor once per month.

o In market surveys, citizens of Santiago (potential users) demonstrated that they increasingly will use Autopista Central.

o The usage of the road is highly correlated with the economy of Chile which has been stable and growing.

ii. Operating Leverage: The project has high fixed costs compared to variable costs and thus represents a higher operational risk. The major costs are depreciation and administrative expenses which is normal in the public infrastructure sector.

iii. Financial Leverage: The project has a highly leveraged structure which increases the risk. The book value of interest bearing liabilities divided by the Stockholders’ Equity between 2006 and 2007 has been higher than 4 (4.9 and 4.2 respectively).

2. Cost of Debt

The cost of the firm’s debt was assumed to be reasonably captured by the stated coupon rates of Autopista Central’s outstanding debt. The company’s debt as of November 2007 was the following:

[pic]

The coupon rates of the above referenced bonds are the following:

• The USD bonds have a coupon rate of 6.223% which was assumed to be the cost of debt in USD. To transform the USD interest rate to Chilean Pesos we used the PPP rule which requires adjusting the rate according to difference in the inflation of the United States[1] and the inflation of Chile[2].

The simplified Power Purchase Parity (PPP) formula assumes that difference in interest rates between countries can be explained by differences in each country’s respective inflation. The PPP formula that was used is the following:

(1+iUS) / (1+InflationUS) = (1+iChile) / (1+InflationChile)

Where:

iUS: USD interest rate

InflationUS: Inflation rate in the United States

iChile: Chilean Peso interest rate

InflationUS: Inflation rate in Chile

Upon applying the PPP formula, we calculated the adjusted interest rate in Chilean Pesos to be 6.540%.

• Bonds in Unidades de Fomento have an interest rate of 5.300% which was assumed to be the cost of the firm’s debt in said currency.

The interest expense of the obligations with related companies is at the cost of equity, which is what we estimated previously.

3. Calculating the Cost of Equity Capital with the Damodaran Modified World CAPM

The calculation of the Damodaran Modified World CAPM (MW-CAPM) is fairly straightforward given that 100% of the revenues of Autopista Central are derived in Chile. That is, the calculation of the Lambda weight that the student is to weight Chile’s Country Risk Premium (CRP) by is:

By using the CRP of Chile that was given in the case (2.10%), the resulting costs of equity capital with this Lambda weighting for the three different project Betas calculated above is:

The resulting costs of equity capital in USD are then translated into costs of equity capital in CLP using the PPP formula described above to obtain the following rates:

4. Calculating the Cost of Equity Capital with the Erb-Harvey-Viskanta Model

The first inputs to the EVH calculator - U.S. risk-free rate, U.S. equity risk premium, U.S. and Chile Institutional Investor Country Credit ratings – should be fairly straightforward as most can be found in the case document itself. In addition, the instructor should remind the class that the EHV calculator is formatted to accept only USD based rates. The rates that we used for the initial inputs were the following:

Once the class is in general agreement about the initial inputs, the instructor should move onto to the discussion of the project risk mitigating factors of the EHV calculator. In this part of the spreadsheet the student must assign a score to each of 11 factors that can increase or reduce the country risk premium that was calculated in the first section of the calculator. Here the conversation should be geared toward incorporating the information given in the section of the case entitled Qualitative and Quantitative Inputs for the Erb-Harvey-Viskanta Cost of Capital Model. The instructor should walk through all eleven factors with the students and try get a consensus for the values to be assigned. Our inputs and corresponding values were the following:

Remind the class of the discussion of the beta that was had earlier and show them in the spreadsheet how a change in this input can change the output of the model.

With the risk inputs above, the EHV cost of capital calculator produces rates in USD for the three different project betas mentioned above. Again, these rates are then translated into costs of equity capital in CLP using the PPP formula described above to obtain the following rates:

5. Damodaran Modified World CAPM WACCs

Using the corporate tax rate, the cost of debt capital, the Damodaran Modified World CAPM cost of equity capital, and the weights of debt and equity in the capital structure, respectively, the WACCs for the three different project betas previously discussed were calculated and can be summarized below:

6. EHV Based WACCs

Using the corporate tax rate, the cost of debt capital, the EHV cost of equity capital, and the weights of debt and equity in the capital structure, respectively, the WACCs for the three different project betas previously discussed were calculated and can be summarized below:

Valuation Results

By using the financial statements included in Exhibit 16 and their accompanying assumptions in Exhibit 17, the students should be able to roughly estimate the unlevered free cash flows of the project. These cash flows, which do not have a terminal value (when the project ends any remaining owner’s equity is dividend back to the owner), are the cash flows that are to be discounted by the WACC derived above. The following graph shows how the Sales, EBITDA, and Unlevered Free Cash Flows of Autopista Central change over the life of the project.

Upon discounting the projects unlevered free cash flows in a manner that accounts for the changing capital structure, we obtained a range of values for the 48% common equity stake of ACS in Autopista Central that went from $524 to 556 million under the Damodaran Modified World CAPM approach and $647 to 684 million using the EHV approach. The following graph shows how the equity value of ACS’ interest in Autopista Central changes with changes in the project beta used in both the Damodaran Modified World CAPM and EHV methods.

How Much Should ACS Ask For?

As shown above, we estimate the range of values for the 48% common equity stake of ACS to be between $647 to 684 million under the Damodaran Modified World CAPM approach and $524 to 556 million using the EHV approach. This range of values is based on WACC values that range between 6.74% and 7.03% when the cost of equity capital is calculated using the Damodaran Modified World CAPM and 7.79% to 8.08% in the case of the EHV methodology. At these valuations, the range of the total value of the 100% common equity of Autopista Central goes from $1.3 to 1.4 billion under the Damodaran Modified World CAPM scenario and $1.1 to 1.2 billion under the EHV scenario. This range of values is significantly higher than the amount of invested capital in the project of $813 million (including subordinated debt).

At this point, the instructor should ask for opinions from the class regarding possible sale prices based on the above mentioned range of values. Who believes that ACS should push for the high end of the price range? Why? Does anyone believe that ACS would be justified in asking for the low end of the price range?

Besides the fact that asking for an initial high price is a good idea in negotiations, we believe that ACS should ask for the highest value obtained in the Damodaran Modified World CAPM model or $685 million. This is recommended because we believe that Abertis and Santander should pay a control premium of 10 – 20% given the macro economic conditions of the time and the strategic value that the Autopista Central asset means for each prospective buyer. If we apply the 20% control premium to the highest range of the EHV based valuation of $556 million, we obtain a value of $667 million. This value is within range that ACS will be pressing to use under the more favorable discount rate that using the Damodaran Modified World CAPM provides them.

Abertis, who is an expert in managing existing toll roads, would presumably be able to generate cost synergies if not revenue synergies when operating Autopista Central. Furthermore, said synergies would probably be easier to realize as the toll road would fit into Abertis’ existing Chilean infrastructure portfolio and probably allow for some reduction in duplicated administrative expenses.

In the case of Santander, finding attractive infrastructure investments that could help them draw more investors to the S2 private equity fund by showing a “winning project” already in the pipeline. A project in Chile, where they are the market leader, must certainly give them a boost of confidence when evaluating the risks and rewards that such an opportunity encompasses. Santander’s high interest in Autopista Central can be easily inferred by the fact that the bank is proactively seeking to buy ACS’ stake rather than the other way around.

Finally, given the exceptionally high prices of real-estate and infrastructure at the time of the case, ACS could easily justify asking for a premium above the value yielded by the DCF model.

What Happened?

In December 2007 ACS agreed to sell both its 48% stake in Autopista Central as well as its 50% ownership in another Chilean toll road to a consortium owned 57.7% by Abertis and 42.3% by Santander (with the goal of Santander being to pass its investment over to the S2 infrastructure fund once said fund had been closed). In addition, the consortium purchased another 2% of Autopista Central’s equity from one of the minority Chilean shareholders. The remaining Chilean minority shareholder sold his 2% stake to Skanska at the same time. As part of the deal Abertis took over operational control of the day to day activities of Autopista Central.

In total, Abertis and Santander paid $1.08 billion (€ 728 million) for the 50% stake in Autopista Central (48% stake from ACS and 2% stake from minority shareholder) and the 50% stake that ACS controlled in the additional toll road. The firms, however, did not disclose the exact amount of the total purchase price that was allocated to each equity purchase. Nevertheless, based on an Abertis shareholders presentation that was issued to announce the deal in December 2007, we estimate that the ACS stake in Autopista Central was valued at just over $620 million. If accurate, this sale price is the falls roughly between the ranges offered by the Damodaran Modified World CAPM and EHV valuation approaches.

[pic]

-----------------------

[1] US Inflation of 2.190% estimated by Aswath Damodaran

[2] Chile Inflation of 7.4% (2006 Autopista Central Financial Statements, Note 2)

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download