The Fuqua School of Business
Emerging Markets Finance
Professor Campbell Harvey
The Privatization of Telebras is one of the largest Mergers & Acquisition (M&A) transactions in
the history of Brazil and Latin America. Telebras was broke-up into 12 separate telephone
service providers, including 3 fixed-line operators, 8 cellular operators and one long-distance
carrier, Embratel. This case focuses on the Privatization of Embratel. From all Telebras's
subsidiaries we chose Embratel because it presents a good example of a company facing a huge
transformation in the markets it competes. Additionally, Embratel's deal sets a perfect ground for
a discussion about risk identification on M&A transactions in Emerging Markets (EM).
This case is set in June 1998, when Deutsche Telekom's (DT) Board of Directors required the
company's M&A team to do a valuation exercise on Embratel before the privatization starts. The
Board has ordered Hans Muller, Managing Director and leader of the M&A team, to accomplish
this task for the next Board's meeting in two weeks. The valuation is going to be used to have an
estimate of the value of Embratel and decide whether DT wants to enter in the bidding process
announced for July 1998. Hans and his team have little time to do the valuation and must justify
the value of Embratel in front of DT's Board. Hans' major concern is how would the fact that
Embratel operates in an EM affect the different valuation methodologies and the information he
has on hand to do this exercise. And whether methodologies used are really affected by
distortions typical of EM and how to adjust them in order to get reasonable results.
The main objective in this case is to set the ground for a discussion on different issues
surrounding a valuation exercise in EM. Specifically, what red flags foreign investors must look
for when doing a deal of this kind in EM? What types of risk foreign investors face in these
markets? The case also examines the use of different valuation methodologies and their efficacy
in a valuation exercise in EM firms.
The case on Embratel's privatization has been written thinking in an advanced corporate finance
course. The case is about valuation in emerging markets and brings in three major topics: risk
assessments, including country risk; discounted cash flows analysis; and comparable analysis.
Although all of these topics have been addressed in the case, some of them have a higher
This case was prepared by Fuqua MBA students Jaime Arriagada ‘99, Luis De Zabala ’99, Fernando Diaz ’99 and
Lorena Navarro ’99 under the supervision of Professor Campbell Harvey (February 1999).
DISCLAIMER: Some of the financial information in this case was estimated for simplicity and some of the events,
organizations and individuals mentioned in it are fictitious.
emphasis. This has been done intentionally in order to conduct the discussion to certain topics
that are not so extensively covered in other cases. The different emphases on the learning
objectives are shown in Exhibit 1.
The objectives of this case are:
1. To have an insight into one of the most important topics in many emerging markets:
privatization of state-owned companies. The case illustrates a real life business situation that
happened in Brazil less than a year ago. The case deals with a situation where the Brazilian
government is looking for modernizing its economic infrastructure. They need to create a
privatization scheme attractive to foreign investors with telecom experience. Additionally
they created a new regulatory framework that is supposed to boost competition and increased
2. To perform a discounted cash flow analysis. In the case there is information so as students
may discuss about some issues such as:
To forecast cash flows using nominal versus real local currency or dollars.
To discuss on relevant time horizon to forecast and to calculate a terminal value.
To determine the key value drivers.
To include in the cash flows other elements such as: taxes or investment credits, legal
3. To perform comparable analysis and compare its results with the DCF analysis.
To determine what are the relevant comparable companies.
To understand country risk involved in every comparable ratio and the appropriateness of
using comparable firm multiple (Comps) from other countries, even in the same region.
To discuss which value ratios are suitable to use in each case (P/E ratios, Value/EBITDA,
To discuss the relevance of current or trailing Comps and price horizons.
4. To discuss how to incorporate country risk, industry risk and company specific risk in the
To discuss the different approach regarding country risk assessment.
To understand the different existent models that incorporate country risk in the discount
To assess the elements composing a discount rate, such as the risk measure with the
market (Beta), the market premium and the risk-free investing alternatives.
5. To select the appropriate currency and time horizon for some key drivers, such as sovereign
debt or market return.
The Case can be divided into Four major aspects:
I. Risk identification in Emerging Markets (EM)
II. Recommended valuation methodologies (DCF and comparable firms)
III. Building the DCF
IV. Searching for comparable firms
Following we present a detailed explanation of these four major aspects.
I. Risk identification in Emerging Markets (EM)
The Case provides material to address the discussion of the risks that any foreign investor is
facing when entering the Brazilian telecommunications industry. A list of the risks mentioned in
the Case is in Exhibit 2.
Some elements to keep in mind when conducting a discussion related to risks in EM are:
To allow students to differentiate among country risk elements, industry risks and the
company specific risk.
Political Risk: Understanding the historical events that happened in Brazil before the Telebras
privatization to assess the country risk. The historical instability and the modernization
process carried over by President Cardoso would be helpful in assessing the likelihood of
different political events that may negatively affect Embratel profitability. Two elements
should be discussed: (i) the privatization process happening in 1998 is the flip side of the
nationalization that occurred in early 70’s, where actually Telebras was created; (ii) telecomm
business are basically based on concession that may be easily revoked if the perceived social
benefits are not fulfilled.
Financial Risks: Student should analyze Embratel clients and suppliers, as well as their
particular bargaining power. That will give an idea of the risks involved in some of the cash
flows. One example of the most important costs for Embratel is the access fee it should pay to
fixed-line companies. The regulation behind this fee as well as the ability of the fixed
telephone companies to press for changes in it should be considered.
Economic Risks: Brazil has a long history of economic instability regarding foreign
exchange, inflation, economic growth, fiscal deficit and so on.
Industry Risks: the private telecom industry in Brazil is new. A state-owned industry for
many years is being transfer to private operator all at once, under a new regulatory
framework. Although many regulations elements seem to be clear and defined, this
regulations are new in Brazil and therefore they will need some adjustments and discussion.
These changes may have effects on Embratel. Additionally, this is an industry that will
gradually increase competition. New competition will bring lower margins and decreasing
Company Risks: Embratel is a company that shows good opportunities to increase efficiency
and profitability. However it may also mean one-time write-off in the Balance Sheet to
adjust to the new competitive environment.
The students should be conducted to analyze the effects of different risks on Embratel
profitability. A sensibility analysis on the different risk elements should accompany the valuation
as well. The discussion should also take into consideration the synergies that different types of
investors may bring to the acquisition, which would affect dramatically Embratel’s value.
II Recommended valuation methodologies (DCF and comparable firms)
Three of the most used methodologies are Discount Cash Flow (DCF) analysis, discount
abnormal earnings, comparable firm multiples valuation and undertaken M&A transaction
multiples of comparable firms. In this Case we are using DCF and comparable multiples
(Comps) because we think these methods address most of the issues we want to discuss.
It is well accepted among financial theorists that the value of a firm should equal to the present
value of the future cash flows1. DCF focuses on discounting cash flows from operations after tax,
after investing in working capital and capital expenditures. Valuation based on DCF can be
structured in two ways. Forecasting cash flows to equity holders and apply a cost of equity
capital to discount such cash flows (plus tax shield of debt). Or forecasting available cash flows
to all providers of capital and apply a Weighted Average Cost of Capital (WACC) as a discount
rate. The second approach is the most widely used because does not require forecasting debt
balances into the future.
Under the Comps valuation approach, one relies on the market to consider the short-term and
long-term prospects for growth and profitability and their implications of the values of
comparable companies. Then it is assumed that the pricing of those firms is applicable to the firm
at hand2. Unfortunately, Comps are not as simple as they appear. Identification of the firms that
are truly comparable to the firm under valuation is very hard. For Embratel the situation is even
harder. With Comps of similar companies in the US and firms in other Emerging Markets, the
multiples vary significantly among those firms. See section "Searching for comparable firms"
ahead in this Teaching Note for detailed explanations on this methodology.
However, both methods have pros & cons. On one hand, DCF analysis incorporates completely
growth potential of the business and can help quantify synergies between the buyer and the target.
Also DCF reflects appropriately the expected return based on the risk of the target company
adjusted by its country risk. On the other hand, DCF is subjective about the operative goals
Embratel would be able to meet. Thus, living the acquirer with capacity to manipulate those
assumptions which leads in some cases to value overestimation. Comps, rather, give a good
value estimation based on a real market value of comparable firms within the same industry.
However, it does not consider the control premium a buyer has to pay to acquire target's voting
control. Additionally, returns expected by marginal investors can be higher than that of an
strategic buyer in the stock market. And finally, the future of some companies that serve as
Comps can be very different from that of the company under evaluation.
III Building the DCF.
The case is rich in financial information aiming to provide the student all the parts to build a full
valuation model and a DCF analysis. Under the DCF approach, students should develop a
projected Free Cash Flow and a terminal value. The exercise starts with projections of the Profit
& Loss statement at least to the level of the operating margin before depreciation (EBITDA). To
explicitly forecast the financial part of the P&L is not necessary for a valuation because the
capital structure is considered in the WACC. Some items of the Balance Sheet Statement need to
"Business Analysis & Valuation", Palepu, Bernard & Healy, 1996
"Business Analysis & Valuation", Palepu, Bernard & Healy, 1996
be projected as well, in order to estimate working capital. The aspects of this DCF analysis can
be subdivided into:
- General assumptions
- Financial projections, Free Cash Flow (FCF) and Terminal Value
- Discounting FCFs (WACC, Cost of Capital and Country Risk)
- Sensitivity analysis and Value at Risk (VaR)
Each one of these subdivisions is explained in detail below.
- General Assumptions:
In order to elaborate a full model and obtain Embratel's value based on DCF it is necessary to set
major assumptions. Those assumptions can be grouped depending on the part of the model they
are intended to rule. Among these assumptions are macroeconomic assumptions, operative
assumptions, market share assumptions and tariff and costs assumptions.
First, it is necessary to specify some macroeconomic assumptions. Our suggested model works
under Purchasing Power Parity (PPP) because we consider that most of the devaluation risks are
incorporated in our discount rate. See Exhibit 6 in the Case. This means that the exchange rate,
and therefore the devaluation rate, is the result of the difference between the inflation in Brazil
and the inflation in the US. PPP3 suggests that the Brazilian currency is going to devaluate as
much as it is necessary to maintain the purchasing power between the two economies. Hence, the
key inputs are the inflation rates for both countries. Projecting the inflation in Brazil and in the
US is out of the scope of this case. Thus we took projections for the US and Brazilian inflation
from a CS First Boston Report4. PPP generates a devaluation rate that allows us to compute end-
of-year (YE) exchange rate (FX) and average exchange rate (AVG FX). Another assumption that
has to be made is the tax rate. Our suggested tax rate is 33%, which is the corporate marginal tax
rate in Brazil. In order to build the P&L it is necessary to project some operative assumptions.
These assumptions are related to the market in general and to the operation of Embratel. The case
provides a set of operative assumptions including total market growth, Embratel's market share
and Embratel's average per minute of usage tariffs.
- Financial projections, FCF and Terminal Value
The projection horizon we suggest is four years. This is because in year 2002 the long-distance
market is going to be wide open to competition and the company would face fierce competition in
the markets it operates. We believe that projecting four years and assessing a conservative
terminal value that accounts for the downturn in Embratel's market share will result in a good
estimate of the value of the company "as is" today. One important issue that might be taken into
consideration is the currency to be used for the projections. Which currency, R$ or US$, in
nominal or in real terms. A simple approach is to make projections in local currency and translate
FCFs into US$ using AVG FX and then apply a US$ denominated discount rate. This way is
easy to simulate effects on the FX and inflation in local currency denominated FCFs. If the
inflation is below 2 digits, projections can be made in nominal terms because the impact of
inflation shouldn’t be dramatic. However, if the inflation is above 2 digits, it is effective to make
projections in real terms. Making projections in real terms may bring some distortions in the
PPP=(1+Brazilian Inflation)/(1+US Inflation)-1
CS First Boston is an Investment Bank that periodically issues research reports on Latin American Stocks.
financial projections. For example, high inflation may decrease revenues and increase costs
therefore reducing margins. The company may be unable to maintain tariffs at the same pace of
inflation while costs are increasing fueled by large inflationary components such as work force
cost. In this Case we are assuming that inflation keeps below 2 digits and therefore no major
considerations are required.
The case includes detailed information on rates per minute of usage in US$ and R$. This
information on both currencies may be relevant to do some sensitivity analysis on FX exposure.
See Exhibit 5 in the Case. With the total market projections in minutes and the market share the
company would have each year of the projection horizon it is possible to compute Embratel's total
minute volume for each year. The resulting volume of usage (in minutes) can be multiply by a
rate per minute of usage in each service and get a revenue per service. Volume and tariff
information for DATA and OTHERS revenues are not given in the case. For simplicity, these
revenues can be projected to grow at the same rate as the total growth rate of DLD and ILD
Once the revenues have been projected, the next step is to estimate the costs of services provided.
The Case includes the access charges fees per minute of usage Embratel has to pay to access local
networks. These fees are given in US$ and R$ in order to encourage sensitivity analyses. With
total minutes of usage and access charges fees is easy to compute access costs. Depreciation has
been historically around 6% of the gross Property, Plant & Equipment amount. For simplicity we
can assume that this ratio is going to persist for the projection period. Personnel and G&A were
4.8% of revenues in the last historical fiscal year (1997). Some Wall Street Analyst Reports
suggest that this is one of the areas where the new owner has room to cut costs. A conservative
approach would be to maintain the 4.8% of revenues. The above computations are enough to
arrive to the cost of services provided (similar to a Cost of Goods Sold in this case). Finally, it is
necessary to estimate other expenses. Selling expenses can be assumed to be the same ratio to
revenues as the last fiscal year. Same situation with Services and Other Operating Expenses.
Now the student can compute the EBITDA for the projected period and the terminal year. Since
EBITDA is more than enough to start a FCF, no financial expense projections are needed.
Regarding Balance Sheet projections only the items those integrate working capital are required.
Specifically, cash and deposits, accounts receivables, accounts payable, inventories, etc. Using
the Financial Statements for 1997' fiscal year, the student can determine the days of revenues for
cash, accounts receivables, etc. Assuming that Embratel would maintain these operative
premises, these items can be backed-out for each year of the projected period. Same
methodology can be used for the accounts payable, but using days of cost of services instead of
revenues. In accounts payable, however, the company's relationships has been mainly with other
Telebras subsidiaries. After the privatization Embratel will no longer be member of the Telebras
holding and advantageous commercial credit would be reduced. A conservative approach would
be to reduce the days of payables from 48 days today to 35 days. See model's assumptions in
Using the Balance Sheet projected items, the student can build working capital needs. The
student now has all the parts to project FCFs. Starting with EBIT, subtracting taxes and adding
back depreciation, requirements in CAPEX and needs in working capital, the FCF can be
computed for each one of the projected years. For CAPEX, a good approach is to use the
estimated commitments in future years, see Exhibit 7 in the Case, and the depreciation expense
for each year and add them up. This is a reasonable proxy of what the CAPEX is going to be.
We add depreciation to it because it represents the CAPEX that is directed to maintenance of the
network rather than extensions of it. The resulting FCFs still are denominated in R$. So the next
step is to translate those FCFs and Terminal Value into US$ using AVG FX rate of each year as
To calculate the FCF in the terminal year we used the same methodology of the previous year but
changing the assumptions to a more conservative level. This change is explained because of
Embratel’s reduced market share after 2002. Then, terminal value in R$ is computed using the
"value driver formula"5 assuming a growth rate of 3%. This Terminal Value in R$ is then
translated to US$ using year 2002 AVG FX rate.
- Discounting FCFs (WACC, Cost of Capital and Country Risk)
The discount rate is one of the main discussion points in this Case. Each time a valuation is done
over a company in EM there are issues around the cost of capital and the discount rate. Students
have to decide whether the discount rate should be in local currency or in US$ and whether this
discount rate has to be in nominal or real terms. Which Betas should be used, comparable
company's stocks versus the S&P 500 or comparable local stocks against local market indexes?
What are the truly comparable firms to the firm at hand? Where should the country risk be
incorporated and how should be estimated? What cost of debt should be used and in what
currency? What capital structure is the ideal for the company at hand? These are few of the
issues involved in any valuation and in particular in a company valuation in EM. The Case sets a
ground for discussion on these issues and this Teaching Note suggests how the instructor should
WACC in US$ or local currency (Nominal Vs Real terms ?):
The instructor has to keep in mind that the discount rate can be denominated in local currency or
in US$. The key consideration here is that both rates have to be consistent with the rest of the
valuation exercise. For example, if the WACC is US$ denominated, FCFs have to be
denominated in the same currency. The cost of debt has to be in US$, the Rf has to be the rate
used in the benchmark country as risk free and the market premium has to be the premium of the
same market with respect to the Rf. If the WACC to be used is in real terms (inflation adjusted)
same considerations apply. FCFs have to be in constant currency, whatever this currency is, cost
of debt has to be inflation adjusted and so on. The main point here is consistency. If students
think that the valuation in local currency terms is more sound there is no problem in that approach
as long as the discount rate is consistent with the cash flows to be discounted. In our Case, we
took the approach of local currency FCFs converted to US$ by each year's FX average rate and
discounted by a US$ nominal WACC.
Cost of Capital under CAPM (Risk Free Rates and Betas)
Using the CAPM6 to determine the cost of capital, students have to come up with a risk free rate,
a market premium and a Beta to plug into the CAPM formula. The Case provides tables with
different risk free rates in the US and many Betas from companies either in the US or in EM. See
Exhibit XXX in the Case. Students have to choose and judge which Rf and which Beta is the
most accurate for Embratel's cost of capital. Our base case approach is to use the "long bond"
Value driver formula is TV=FCF/(WACC-g)
Capital Asset Pricing Model is CAPM=Rf+*(Rm-Rf)
(30-Y Treasuries) as the Rf. The market premium is assumed to be 6.21%7. And for the Beta we
took the average of all Betas provided (after re-levered using Embratel’s capital structure).
Cost of Capital and Country Risk (Company Risk and Country Risk)
The Case poses a lot of emphasis in the way default risks should be taken into consideration and
how those risks are actually estimated. A valuation exercise can be defined as the value obtained
by discounting a stream of expected cash flows at a selected discount rate. Now this exercise
includes two parts. The Numerator, or the stream of cash flows and the Denominator, or the
discount rate (WACC). Some financial theorists argue that a country risk adjustment must be
done over the cost of capital affecting therefore the discount rate. This adjustment changes the
denominator of the equation, which in EM cases drives the cost of capital up and along with it,
the discount rate. The effect is that the company's value would be lower because we are adjusting
for a higher Country Risk. Other practitioners argue that the probabilities of each risk factor
should be incorporated in the cash flows themselves (e.g., FX rates and inflation).
Our approach is that the denominator should be adjusted to account for the Country Risk (and
there are several methodologies to do it) and some sensitivities must be done in the numerator for
certain key FCF drivers. Doing this, the Country Risk adjustment in the WACC accounts mostly
for risks difficult to make sensitivities in the FCFs, such as civil wars, expropriation, and
deterioration of the Country's overall economy among others. Some Country Risk issues that can
be sensitized through FCFs are FX rates, inflation and changes in Embratel’s operations due to
worsened overall economic situation (such as prices and costs and working capital investments).
Two methodologies to determine Cost of Capital adjusted by Country Risk (Spread Model and
There is no question about whether the discount rate should include an adjustment for Country
Risk. It has to become clear to the students that a discount rate used in the US cannot be the same
to a discount rate used in Brazil. In this Case we explore two methodologies to incorporate the
Country Risk in the Cost of Capital. Following is an extract from Professor Harvey's paper on
International Cost of Capital. Later we review the Spread Model and explore Professor Harvey's
IICCRC International Cost of Capital and Risk Calculator.
"One might consider measuring systematic risk the same way in
emerging markets as well as developed markets. Harvey's (1995) study of
emerging markets returns suggests that there is no relation between
expected returns and betas measured with respect to the world market
portfolio. A regression of average returns on average betas produces an R-
square of zero. Harvey documents that the country variance does a better
job of explaining the cross-sectional variation in expected returns."
"Indeed, the evidence in Harvey (1995) shows that, over the 1985-
1992 period, the pricing errors are positive in every country in the IFC
database. This implies that the model is predicting too low of an expected
return in each country. In other words, the risk exposure as measured by
the world model is too low to be consistent with the average returns."8
"The International Cost of Capital and Risk Calculator (ICCRC)", Campbell Harvey, 1995
Professor Harvey's point is that some Country Risk calculators are somehow static and tend to
measure this risk at one point in time, without taking into consideration the country’s previous
volatility. In the case of Brazil, the country holds a strong history of economic distortions and
represents a great risk. Following is the first method we used in the Case, The Country Spread
Model, defined by Professor Harvey.
"When regressing the company returns (measured in US dollars) on
the benchmark return (either US portfolio of the world portfolio), the Beta
is either indistinguishable from zero or negative. Given the correlation
between many of the emerging markets and the developed markets are low
and given the evidence in Harvey (1995), it is no surprise that the
regression coefficients (Betas) are small. The implication is that the cost of
capital is the US risk free rate or lower."
"The following is a popular modification used by a number of
prominent investment banks and consulting firms. A regression is run of
the individual stock return on the S&P 500 stock price index return. The
Beta is multiplied by the expected premium on the S&P 500. Finally, an
additional factor is added which is sometimes called the "Country Spread".
The spread between the country's government bond yield for bonds
denominated in US dollars and the US Treasury bond yield is added in.
The bond spread serves to increase an unreasonably low cost of capital into
a number more palpable to investment managers."
"There are many problems with this type of model. First, the
additional factor is the same for every security. Second, this factor is only
available for countries whose governments issue bonds in US dollars.
Finally, there is no economic interpretation to this additional factor. In
some way, the bond yield spread represents an ex ante assessment of the
country risk premium, which reflects the credit worthiness of the
government. However, beyond this, it is difficult to know how to fit this
factor into a cost of capital equation"9
We would add to this analysis that in the Spread Model a country's bonds may be not liquid,
which may also affects their prices and thus their YTM10. Also, bond spreads are a measure at
one point in time and does not account for prior volatility history of a country. In Figure 1, we
see that Brazil's Brady bonds have been at higher prices in the past. This suggests that Brazil is a
country with marked economic distortions even though the Real Plan seems to be functioning.
Anyway, the situation might turn around because Brazil's economy is weak, and little instability
may reduce bond prices even further, increasing their YTM and thus their spread. This valuation
is done in June 1998, the lowest spread of Brazilian Brady Bonds with respect to the US
Treasuries since 1994. (See Figure 1). Is this a good representation of the risk DT is assuming by
acquiring Embratel? We think if this exercise had been done in early 1996, Embratel’s value
would have been very different. Is this reasonable?
"The International Cost of Capital and Risk Calculator (ICCRC)", Campbell Harvey, 1995
YTM= Yield to Maturity
Brazil Brady Bond Spread over 30yr. US Treasury
The second methodology we explored in the Case to incorporate Brazil's Country Risk was Erb-
Harvey-Viskanta's (1996) ICCRC11. Harvey defines the model as following.
"This model specifies an external ex ante risk measure. Erb,
Harvey and Viskanta (1996) require the candidate risk measure to be
available for all 135 countries and available in a timely fashion. This
eliminates risk measures based solely on the equity market. This
also eliminates measures based on macroeconomic data that is
subject to irregular releases and often dramatic revisions. They focus
on country credit ratings."
"The country credit ratings source is the Institutional
Investor's semi-annual survey of bankers. Institutional Investor has
published this survey in its March and September issues every year
since 1979. The survey represents the responses of 75-100 bankers.
Respondents rate each country on a scale of 0 to 100, with 100
representing the smallest risk of default."
"One dimension of this analysis is the estimation of sovereign
credit risk. The higher the perceived credit risk of a borrower's home
country, the higher the rate of interest that the borrower will have to
"The International Cost of Capital and Risk Calculator (ICCRC)", Campbell Harvey, 1995
pay. There are many factors that simultaneously influence a country
credit rating: political and other expropriation risk, inflation,
exchange rate volatility, and its sensitivity to global economic
shocks, to name some of the most important."
"The credit rating, because it is survey based, may proxy for
many of these fundamental risks. Through time, the importance of
each of these fundamental components may vary. Most importantly,
lenders are concerned with future risk. In contrast to traditional
measurement methodologies which look back in history, a credit
rating is forward looking."
In the Case, the protagonist Hans Muller is a former Duke MBA student who contacts Professor
Harvey at Duke's Fuqua School of Business. Professor Harvey provides Hans with the outcome
of the ICCRC for several Betas the student wants to have in order to calculate the cost of capital
for Brazil. The student, however, must make a judgement call on what Beta should be plugged
into the ICCRC calculator. Our based case assumed an average Beta (1.13) of all the Betas
provided, after un-levering and re-levering those using Embratel's long-term D/V.
Cost of Debt
The Case gives students with an estimated long-term average cost of debt of 10.5% in US$.
Students may backed-out financial expenses from the P&L and the Balance Sheet for fiscal year
1997. This is a reasonable approach although is not forward looking. Is better if the student uses
the 10.5% provided.
Long Term Capital Structure (D/V)
We provide the student the long-term capital structure to be 30% Debt to Value, to be used in the
WACC. This ratio comes from analysts’ reports.12
Putting together all the pieces, the student can build a WACC to be used for Embratel. Our base
case shows a WACC using CAPM's cost of capital with the Country Spread Model. Also a
sensitivity table is provided in this Teaching Note using ICCRC. The instructor can tell the
difference between both methods, which is in-line with what is explained above. See Exhibit 3
and Exhibit 4. Also see Full DCF Valuation Model at the end of this Teaching Note.
- Sensitivity analysis and Value at Risk (VaR)
To be consistent with the numerator-denominator theory exposed before in this Teaching Note,
we decided to incorporate country default risk in the WACC (through cost of capital), and adjust
FCFs (numerator) to reflect company specific probabilities of bankruptcy. The student can
perform several sensitivity analyses first on the WACC (through cost of capital, perpetuity
growth rate and country risk) and later over key FCFs drivers, such as FX rates, market share of
Embratel, telephone service tariffs and interconnection rates among others. By doing this the
student can apply the concept of Value at Risk (VaR). Trends can be estimated on these specific
Latin America Investment Research, Morgan Stanley Dean Witter, June 2, 1998
drivers looking at their history and at some subjective information given in the Case. Various
scenarios can be simulated and probabilities of those scenarios can be estimated.
IV. Searching for Comparable Firms
In this section there are three main issues that the students should address:
- What multiple should they use? (e.g. P/E, EV/EBITDA, EV/LIS)
- Which companies might be considered comparable?
- Should they incorporate country risk in the analysis? If so, how?
The case has enough information to build the most commonly used multiples. Among those than
can be calculated are: P/E, EV/EBITDA Estimated, P/Cash Earnings, P/BV, EV/Revenues.
Although practitioners use some of them regularly, we think that EV/ EBITDA Estimated is the
most rigorous among them. Enterprise Value takes into consideration the difference in leverage,
while the EBITDA is a good proxy for cash flows and eliminates potential distortions due to
difference in accounting practices for depreciation. Students should identify that they need
'Estimated EBITDA' multiples, rather than trailing ones, to incorporate the available information.
Some students might ask what stock price should they use for the analysis, if the current price, or
some kind of average. Based on financial theory, they should use the current stock price, since it
incorporates all the available information at that time. Nevertheless, they should perform
sensitivity analysis to understand how changes in stock prices will impact the multiple. An
example of all the calculation is shown in Exhibit 5.
To determine which companies are comparable, the case provides a description of each company.
The students should be able to notice that the Latin American telecom companies are not
comparable at all with Embratel because most of them integrate local, long distance and wireless
telecommunications, while Embratel just provides long-distance services. Similarly, the Telebras
multiple itself can not be applied without adjustments to Embratel, because they have different
growth prospects. However, the multiples of the "Latin comps" can be useful to understand how
to incorporate country risk in the analysis. In our opinion the true "comparable" company in the
case is Sprint, although some adjustments are necessary. Sprint is kind of a "pure" long-distance
company, similar of what Embratel was supposed to be.
With the information provided about the importance of the telecom companies in Latin America
and the high correlation with their respective local markets, the students should be able to identify
that some kind of country risk adjustment should be applied to multiples. See Exhibit 6.
Although 'how' to adjust for country risk in analysis of multiples has not been formally addressed
in academic papers, we tested the hypothesis that if Institutional Country Credit Ratings has a
good level of predictability for market returns (Harvey, 1995), it should also explain part of the
difference in multiples. Therefore, we calculated the EV/EBITDA 98E for each of the Latin
telecoms, and using their relative differences in country ratings, we tried to predict the Telebras
multiple. The results were quite interesting results (see Exhibit 7). In this case, we improved our
ability to adjust multiples for country risk. Based on that approach, we estimated the Embratel
multiple adjusting the Sprint multiple to its relative credit rating. We obtained a multiple for
Embratel of 3.3x EV/EBITDA 98E, which was very similar to that estimated by Wall Street
Once the multiple is determined, the next challenge is to estimate the EBITDA 98E figure in US$
(since the available information for the comps is in US$). In this issue, the student is expected to
do some sensitivity analysis to calculate a "normalized" EBITDA and an exchange rate to
translate de Reais EBITDA figures. A referential valuation and its sensitivities can be found in
Exhibit 8. Some students can use this multiple to calculate the terminal value for the DCF. In
our opinion, a number significant higher than 3.3x for the exit multiple can lead to overvaluation.
In this case, the two positions of bidding or not bidding for Embratel can be well supported due to
the potential difference in discount rate and/or main assumptions. Exhibit 9 shows how with a
base case valuation, none of the method have an overlapping range of value. However, we
should expect from those students with a non-bidding position, they should have some kind of
key drivers sensitivity to understand under what circumstances they would bid. This can be an
interesting discussion about if the bidder should be willing to pay part of their synergies for the
target (e.g. increase operating efficiencies, improve revenues per employee, etc). For those
students that are willing to bid, they should state under what circumstances they would not do it.
In July 1998, MCI Communication Corp, the second largest U.S. long distance company, paid
2.65 billion Reais (US$ 2.26 billion) for Embratel. The bid was 47% over the government's
asking price of 1.8 billion Reais. Sprint, the No. 3 U.S. long-distance provider, offered the best
price in the sealed enveloped, R$ 2.499 billion versus R$2.477 billion offered by MCI. However,
because this difference was less than 5%, the auction went to open outcry. MCI acquired
Embratel in less than a minute for R$2.65 billion.
As Bloomberg comments, "the purchase of a 52-percent voting stake in Embratel gives MCI
more customer in the fast-growing Latin America market as it races against AT&T Corp. and
others to serve multinational corporations".
In the last quarter of 1998, the Asian financial turmoil spread, dragging down Russia, and in
January 1999, Brazil. The real plunged more than 40%. As "The economist" said13, "the
depreciation (of the real) creates a serious danger of higher inflation; the importance of defeating
expectations of rising prices was what lead the government to hang itself on an exchange-rate peg
in the first time. If inflation is not to take off again, the currency must be stabilized, which in turn
implies a period of high interest rate. Hence the dilemma: if the central bank keeps interest rates
up to buoy the Real, the burden of servicing the government's domestic debts may become
To prevent a return to hyperinflation, Brazil had tried to increase its fiscal and monetary
discipline without that much success. Mr. Cardoso is no longer the popular leader he once was.
He is trying to make austerity a priority. On March 1999, he will send a tax reform and a "fiscal
"Brazil's slippery slope", The Economist February 6 th 1999
responsibility" bill for congressional approval, which is expected to set limits to spending. More
negotiations are expected.
From the case's learning perspective what is very interesting is to evaluate the Brazilian sovereign
bond spread after the crisis. Exhibit 10 shows that spread over time. The question that can be
posted to students is: what would be the value of Embratel under this new information?