[PPT] - Emerging Capital Markets AG907 M.Sc. Investment & Finance PowerPoint Presentation

SLIDE 1

Emerging Capital Markets AG907

M.Sc. Investment & Finance M.Sc. International Banking & Finance Lecture 4 – Valuation in Emerging Capital Markets

I g n a c i o R e q u e j o – G l a s g o w , 2 0 1 0 / 2 0 1 1

SLIDE 2

Overview of Lecture

1. Introduction
2. Basic Valuation Approaches
3. Valuation Process
4. Does the Choice of Model Matter?

Empirical Evidence – Valuation in Emerging Markets

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

1

SLIDE 3

1. Introduction

Characteristics of EMs affect valuation assumptions ⇒ Adjustment in valuation methods Given foreign currencies ⇒ Two different general valuation approaches Valuation of EM investments ⇒ Steps investors should follow

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

2

SLIDE 4

2. Basic Valuation Approaches

Two discounted cash flow (DCF) valuation approaches for valuation of investments in foreign country ⇒ Example Approach A.

Peso flows converted to dollars (forecast of forward

peso/dollar exchange rates)

Dollar CFs discounted using dollar weighted-average

cost of capital (WACC)

Estimation of dollar WACC must reflect
Systematic risk of target industry
Local equity market & country risk

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

3

SLIDE 5

2. Basic Valuation Approaches

Approach B.

Local CFs discounted using local WACC ⇒ Local NPV

for investment

Local NPV converted into dollar NPV using spot

peso/dollar exchange rate Each approach has Strengths & Weaknesses ⇒ choice

Relative confidence in local capital market data
Relative

confidence in theoretical equilibrium in international currency & capital markets

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

4

SLIDE 6

2. Basic Valuation Approaches

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

5

Peso CFs US Dollar CFs Peso NPV US Dollar NPV

Convert CF from Pesos to Dollars

Approach A Approach B

Discount at US Dollar WACC Discount at Peso WACC Convert NPV from Pesos to US Dollars

SLIDE 7

2. Basic Valuation Approaches

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

6

Characteristic Approach A Approach B

Key features

PPP & IRP hold
Inflation forecasts in pesos &

dollars appropriate

Country risk premium estimate

appropriate

Local capital market has good

data availability & quality

Local capital costs are free-market

yields Strengths

Theoretical rigour
More reliable capital market

information from DMs

Simplicity
Translation at current spot rates

Weaknesses

PPP & IRP do not always hold
Long-term forecasting of inflation

difficult

Assumes US interest rates

consistent with forward peso/dollar exchange rates

Availability & quality of local data
Betas not available for many

stocks in EMs

Many interest rates administered

by central banks & do not reflect inflation expectations or require real rates of return

SLIDE 8

2. Basic Valuation Approaches

Approach A.

Future CFs converted from foreign to home currency

before discounting

Forward exchange rates beyond 3 years rare! ⇒ Rely
n exchange rate forecasts or interest rate parity (IRP)
IRP ⇒ No arbitrage in currency markets; fluctuations in

exchange rates depend on ratio of local & foreign inflation rates Approach B.

Use of local market data to use a local discount rate
Confidence in local data to estimate local NPV!

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

7

SLIDE 9

3. Valuation Process

Three separate steps

1. Evaluate general investment environment
2. Forecast CFs of investment
3. Estimate cost of capital to discount CFs

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

8

SLIDE 10

3.1. General Investment Environment

Intangible attributes of particular market that modify risk of investing in it Investors should consider five key attributes

Information environment
Market integration
Political risk
Rule
f

law, corruption, corporate governance & minority shareholders protection

Social issues & culture

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

9

SLIDE 11

3.1. General Investment Environment

Information environment

Valuation of companies ⇒ Availability of accurate &

reliable information

Data sources for foreign investors ⇒ Accounting

statements & financial analysts projections

Quality of information environment in which company
perates
Investors must adapt to information available for each

market

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

10

SLIDE 12

3.1. General Investment Environment

Market integration/segmentation

Local

market integration with global markets is important ⇒ Fundamental assumption in valuation: that investors can engage in arbitrage

If arbitrage is not possible ⇒ Reference point for

investors is local (rather than global) cost of capital

Integration affects discount rate used in EMs

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

11

SLIDE 13

3.1. General Investment Environment

Political risk

Local government intervention ⇒ Effect on asset value
Different ways of intervention ⇒ Regulation, punitive

tax policies, restrictions on cash transfers, etc.

Outright expropriation of assets of foreign companies
vs. lack of protection when breakdown in order occurs
Official corruption ⇒ Alternative form of taxation
Investor’s perception about political risk ⇒ Estimation
f CFs & exchange rate (CPI)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

12

SLIDE 14

3.1. General Investment Environment

Rule of law, corruption, corporate governance & minority shareholders protection

Valuation depends on protection of minority investors

rights ⇒ Lower share prices if expropriation expected

Premiums for control will tend to be higher ⇒ Access

to higher private rents (private benefits of control)

More corrupt environment in EMs than DMs

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

13

SLIDE 15

3.1. General Investment Environment

Social issues & culture

Some business cultures endorse certain principles
Nepotism
Paternalism
Discrimination
Charitable giving
Official corruption
Tax evasion
Reliance on government assistance

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

14

SLIDE 16

3.1. General Investment Environment

Adjustment in valuation to account for EMs attributes ⇒ Mutually exclusive to avoid double-counting

1. Adjust CFs downward
2. Adjust discount rate upward

Comparing bond yields ⇒ Bondholders premium

Bonds denominated in same currency (US dollars)
Similar class of asset (e.g., sovereign bonds of the

respective government, same corporate bond rating)

‘Free market’ yields (not s.t. government intervention)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

15

SLIDE 17

3.2. Estimating Cash Flows

Important considerations when estimating CFs in EMs

Inflation
Foreign currency exchange rates
Tax rates
Timing of remittance of cash
Accounting principles

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

16

SLIDE 18

3.2. Estimating Cash Flows

Inflation

Wide variation in inflation rates across countries ⇒

applying inflation rate of US, OECD countries, world economy!

Inflation assumptions for CFs & discount rate
Consistent assumptions through valuation process

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

17

SLIDE 19

3.2. Estimating Cash Flows

Foreign currency exchange rates

Significant

variation during investment lifetime ⇒ Forecast exchange rate behaviour

For long-term investments ⇒ Use IRP coupled with

inflation expectations (to estimate future exchange rate behaviour if freedom to move capital across borders)

No freedom ⇒ Take into account capital controls, other

macroeconomic policies that prevent capital from flowing to most profitable opportunities

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

18

SLIDE 20

3.2. Estimating Cash Flows

Tax rates

Affects both forecasted CFs & discount rate
Consistency through valuation process
Marginal tax rate appropriate to country where CFs are

generated

Consider tax system of investor’s country ⇒ E.g., are

investors exempt from taxation of foreign income?

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

19

SLIDE 21

3.2. Estimating Cash Flows

Timing of remittance of cash

Possibility of no free movement of capital across

borders ⇒ E.g., limits in outbound movement of cash

Timing of remittance of cash can affect taxation ⇒

E.g., some countries tax foreign income when it is received (rather than earned)

Timing of the return of cash should be explicitly

modelled

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

20

SLIDE 22

3.2. Estimating Cash Flows

Accounting principles

Deriving CFs from financial statements ⇒ Familiarity

with accounting principles in foreign country

Five different regional profiles of principles
Anglo-American-Dutch
Continental European
South American
Islamic
Mixed economy
System’s particularities will affect forecasting of CFs

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

21

SLIDE 23

3.3. Estimating Discount Rate

Step related to asset-pricing theory ⇒ WACC: the cost
f equity
General principle ⇒ estimate cost of equity consistent

with company risk in EM

Classic errors
Using DM discount rate for EM investments
Using

discount rates inappropriate for target company’s business risk

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

22

SLIDE 24

3.3. Estimating Discount Rate

An overview of models

Several asset pricing models

can be applied ⇒ Different assumptions about pricing of securities & differing demands on data quality & computational sophistication

Models divided according to two key attributes
1. Quality of information
2. Market segmentation / integration

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

23

SLIDE 25

3.3. Estimating Discount Rate

Questions about information in three areas

Market data ⇒ Do security prices reflect what is known

about companies? Is local market relatively efficient?

Company-specific information ⇒ Is financial reporting

transparent & reliable?

Availability & quantity of information ⇒ Long enough

time series? Index dominated by one / two securities = adequate proxy for true market portfolio?

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

24

SLIDE 26

3.3. Estimating Discount Rate

Availability of information: Points to be considered

Market liquidity ⇒ Prices observed in liquid market are

more reliable (in illiquid markets over- or under- valuations cannot be arbitraged rapidly)

Concentration ratios in market indices ⇒ Market index

dominated by few companies will not necessarily reflect performance of market as a whole

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

25

SLIDE 27

3.3. Estimating Discount Rate

Availability of information: Points to be considered

Government intervention & regulation ⇒ Governments

can prevent capital movements (= lack of liquidity); can intervene as large buyer or seller

Market efficiency ⇒ Prices might not reflect relevant

information (EMs do not exhibit weak form efficiency); prices cannot be used to determine expected returns

r industry betas

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

26

SLIDE 28

3.3. Estimating Discount Rate

Market integration/segmentation: As with information, degree

f

market segmentation needs different approaches to valuation

If local market is integrated, its risk exposure will differ

from segmented market

One should consider different asset-pricing models

according to level of integration

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

27

SLIDE 29

3.3. Estimating Discount Rate

Possible situations when valuing assets in EMs

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

28

Information environment Target country integrated Target country segmented Foreign capital market information easily

btained & believed to be reliable (e.g., foreign

capital market relatively competitive & efficient, & financial performance reporting relatively transparent & reliable)

A B

Foreign capital market information not easily

btained &/or unreliable (e.g., foreign capital

market relatively less competitive & inefficient, &/or financial performance reporting relatively

paque & unreliable)

C D

SLIDE 30

3.3. Estimating Discount Rate

A. Asset

pricing is globally integrated and quality information may be obtained

Large, multinational companies actively traded & listed

for trading in developed country exchanges

Companies give financial reports & disclose corporate

news (consistent with DM standards)

Assumption that securities of these companies are

priced without segmentation

In this quadrant, investor can estimate parameters
Companies that operate in EMs on ‘borderline’ to

become DMs (South Korea)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

29

SLIDE 31

3.3. Estimating Discount Rate

B. Where asset pricing is segmented and quality of

information may be obtained

Large companies actively traded in country exchanges

for which segmentation effects of EMs are expected

Companies integrated into global product markets,

followed by numerous securities analysts, traded regularly (e.g. large Brazilian oil company Petrobras)

Local equity markets not highly efficient, (security

prices reflect what is known!)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

30

SLIDE 32

3.3. Estimating Discount Rate

C. Where asset pricing is globally integrated but foreign

capital market information is questionable or difficult to

btain
Assumption

that target is globally integrated but investor cannot obtain reliable data

Models in this quadrant estimate required returns

based on benchmarks from outside EM & company

Should we assume existence of globally integrated

market without reliable market data?

Exceptional cases!

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

31

SLIDE 33

3.3. Estimating Discount Rate

D. Where asset pricing is segmented and information is

questionable or difficult to obtain

All new enterprises, joint ventures, project financings &

foreign direct investments in segmented markets

Companies in EMs far from being classified as DMs

fall into this category

Data problems are severe

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

32

SLIDE 34

3.3. Estimating Discount Rate

Main conclusions when choosing asset-pricing model

One size does not fit all ⇒ Analysing alternative

models & circumstances in which they are useful

Five general types of models can be used in EMs
CAPM
International CAPM
CAPM adjusted for political risk & segmentation
Multifactor models
Country credit model

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

33

SLIDE 35

3.3. Estimating Discount Rate

Main conclusions when choosing asset-pricing model

The choice depends on:
Investor’s perception of quality of information
Segmentation between EM & investor local market
For EMs integrated with global markets & quality of

information ⇒ CAPM, ICAPM, multifactor models

For EMs not integrated with home market ⇒ CAPM

adjusted for political risk & segmentation

For EMs with no quality information ⇒ Credit model

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

34

SLIDE 36

4. Does the Choice of Model Matter?
Choice of model in DMs is less significant
Evidence for five EMs (BRA, SA, THAI, MAL, POL)
Models ⇒ CAPM, CAPM adjusted for political risk &

segmentation, ICAPM, multifactor method

In some cases, similar costs of capital, but in general

differences in estimates (from 300 to 1,000 bps)

Reasons

⇒ Alternative beta estimates, inflation, political risk & equity market returns

Bottom line ⇒ Cost of equity differences are large

among EMs & among models

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

35

SLIDE 37

‘Valuation in Emerging Markets: A Simulation Approach’

García-Sánchez, J., Preve, L. & Sarria-Allende, V. Journal of Applied Corporate Finance 22 (2010) 100–108

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

36

SLIDE 38

Starting Point

Problems to use standard valuation techniques in EMs

⇒ Differences: transparency, liquidity, governance, transaction costs, volatility

Important difference ⇒ Probability of crisis events

(e.g., during the 1990s in Russia, Southeast Asia, Latin America)

Previous approaches ⇒ Adjust discount rate
Inability to use local data to obtain direct estimate
f discount rate
Lack of empirical support for integration

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

37

SLIDE 39

Starting Point

DCF technique ⇒ Discount unconditional expected

CFs at discount rate that reflects symmetric risk not hedged with globally diversified portfolio

Truly unconditional ⇒ Consider potential countrywide

crises & costs of financial distress

Substantial
verestimation
f

value from using conditional expected CFs

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

38

SLIDE 40

Estimation of the Cost of Capital

Cost of equity using CAPM ⇒ Ke = Rf + β x MRP
Rf = risk-free rate
β = systematic risk
MRP = market risk premium
To estimate parameters ⇒ Local market (EM) or global

market (GM) information

EMs information available & reliable?
Level of market integration?
Intermediate solution ⇒ Use of GM data & adjustment

for typical sources of risk in EMs

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

39

SLIDE 41

Suggested Adjustments to CAPM

CAPM ⇒ Non-diversifiable (market) risk
Other sources of risk in EMs? ⇒ Country risk: e.g.,

expropriation by local government, unstable rule of law, lack of transparency

Measure of CR ⇒ Sovereign bond spread
Add measure of country risk (CR) to global CAPM

formulation ⇒ Ke = Rf + βind_us x MRP + CR

Rf = US risk-free rate
β = Beta for comparable firm/industry in US
MRP = US market risk premium
CR = Country risk (spread of sovereign bonds)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

40

SLIDE 42

Suggested Adjustments to CAPM

What are implicit assumptions? ⇒ DCF techniques

imply unconditional expected CFs & corresponding risk

E.g., CAPM & perpetual CFs
In EM context ⇒ Impact of economy-wide crisis?

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

41

MRP R ExpectedCF Value

f

× + = β CR MRP R ents NoCrisisEv ExpectedCF Value

f

+ × + = β |

SLIDE 43

Suggested Adjustments to CAPM

Adjustment violates CAPM theoretical framework
Discount conditional expected CFs & add CR to

discount rate (to account for probability of EM crisis)

Another problem ⇒ Assumption of same impact for all

businesses (level of connectedness to foreign markets, exposure to commodity prices, degree of regulation)

Firm operating only in domestic economy
Firm directed towards international markets

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

42

SLIDE 44

The Country Risk Adjustment

Main differences between investments in DMs & EMs

Higher risks in EMs ⇒ Volatility risk plus political, legal,

social & economic issues

Application of standard methods not appropriate ⇒

Adaptation = add CR to discount rate

Oversimplification ⇒ Assumption that CR
Same for all industries & companies
Represented accurately by sovereign bond spread

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

43

SLIDE 45

The Country Risk Adjustment

CAPM discount rate includes non-diversifiable market

risk ⇒ Symmetric

Country risk reflects CF shortages due to economic,

politic, social, legal conflicts (in EMs) ⇒ Asymmetric

Violation of foundations of basic model
Combination of components of different nature into

discount rate

Estimation of not truly expected CFs

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

44

SLIDE 46

The Country Risk Adjustment

Symmetric vs. Asymmetric Risk Factors

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

45

SLIDE 47

The Country Risk Adjustment

New proposal ⇒ consider impact of CR in estimated

CFs (not discount rate)

Estimation of unconditional expected CFs discounted

at global discount rates

Possibility to allow for different impact of CR for

different companies/industries

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

46

) , , ( ] | [ k CountryRis MarketRisk

ney

TimeValueM f te DiscountRa NoCrisis CF E NPV = = ) , ( ] [ MarketRisk

ney

TimeValueM f te DiscountRa CF E NPV = =

SLIDE 48

Concept of Country Risk

Country risk premium ⇒ Sovereign bond spread CR of shareholders = CR sovereign bondholders Bond holdings

Discount promised CFs at risk-free rate (US) or at

discount rate that includes default premium (EM)

Two components of default premium
Default probability
Recovery rate

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

47

SLIDE 49

Concept of Country Risk

Payoffs from US & EM Bonds

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

48

SLIDE 50

Concept of Country Risk

Equity holdings

US ⇒ Discount most probable at market risk adjusted

discount rate

EM ⇒ Discount conditional expected CFs at market &

country risk adjusted discount rate Why default premium on sovereign bonds = incremental equity risk in EM companies?

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

49

SLIDE 51

Concept of Country Risk

Discounting Conditional Expected CFs

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

50

SLIDE 52

Concept of Country Risk

Key components behind EM’s sovereign bond spreads

Default probability of government
Recovery rate expected by bondholders

Sovereign bond spread = CR ⇒ Two assumptions

Both components important for assessing equity risk ⇒

Default prob. of government = prob. economic crisis

Recovery rate for bondholders = Recovery rate for

shareholders

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

51

SLIDE 53

Concept of Country Risk

EMs bondholders recovery rates reliable indicator of shareholders recovery in crisis times? In crisis, value of companies will vary depending on

Fundamental of business
Degree of government intervention

E.g., in Argentina 2002 ⇒ Firms with significant exports

vs. regulated businesses in domestic market

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

52

SLIDE 54

A Simulation Approach

Solution proposed ⇒ Capture effect of CR not in discount rate but in estimation of CFs Estimation of unconditional expected CFs ⇒ Two inputs

1. Probability of crisis ⇒ Estimate using default prob.

implicit in sovereign bond pricing

2. Recovery

values expected by shareholders ⇒ Fundamentals of the business & degree of government intervention After estimating unconditional expected CFs ⇒ discount rate with only market volatility risk (Global CAPM)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

53

SLIDE 55

A Simulation Approach

Estimating unconditional expected CFs

Estate
f

EM = random variable with binomial distribution (p = default probability)

A company’s recovery value = f (analyst’s views &

assumptions) Discount rate does not include CR component Final

utput

⇒ Estimated NPVs represented by ‘histogram’ of valuations (sensitivity analyses)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

54

SLIDE 56

A Simulation Approach

Three main benefits

Avoid use of conditional expected values & ad hoc

discount rates

Incorporate firm data about recovery values (instead of

EMs bond spreads)

Better analytical tools ⇒ Estimated distribution of NPV

values

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

55

SLIDE 57