Risk Premium


The risk premium corresponds to excess financial return, i.e. it represents the difference between the risk of investing in the stock market and investing in risk-free assets.

Risk Premium = Rm - Rf

Where:

Rm - expected return from the stock market
Rf - risk-free interest rate

PTC has been using information from Bloomberg to calculate the value of the risk premium.

Both in terms of methodology and in conceptual terms, the determination of the expected risk premium of the stock market is not a task which generates consensus. The analysis of risk premium behaviour, however, is complicated because, along with its determinants, it is not directly observable. In addition, risk premium varies over time according to how investors perceive the risk of the asset in question and also according to their attitude towards risk.

In fact, risk premium is an extremely volatile variable, whereby emphasis should be given to observations based on longer data series, as advocated by Damodaran in the study on the risk premiums 1, where it is stated that, because they provide for a smaller standard error, observations based on the longest data series outweigh the benefit derived from the more relevant observations associated with shorter and more recent periods. In this sense, the result becomes more consistent as the period considered is lengthened.

However, taking into account the extent to which the Portuguese market is integrated in international financial markets and the fact that the data set available for the market itself is limited, the Benchmark of more mature financial markets may constitute a good reference for the risk premium of the market because they provide for a longer historical period and therefore for a reduction in the estimation error.

Assuming, therefore, that the Benchmark proves to be an appropriate solution, it is fundamental to define the methodology with respect to: (i) use of ex-post data versus ex-ante data; and (ii) consideration of the arithmetic average of the data series versus the geometric average of the data series.

In general, the ex-post methodology consists of estimates based on historical data, in contrast to an ex-ante methodology which takes account of estimates based on future expectations. Furthermore, the geometric average corresponds to the average annual return with respect to an investor who buys and maintains their portfolio with a long-term perspective, while the arithmetic average reflects the average annual return associated with an investor who makes a daily decision to invest and has no objective to hold the company's capital over an indefinite and extended period.

Ex-post versus ex-ante methodology

Current practice for estimating the market premium is the use of premiums made and observed over long historical series (25 to 100 years), as explained in PwC's report 2. Furthermore, according to PwC's analysis 3, it is evident that most countries favour an ex-post methodology.

However, this premium (calculated from ex-post data) may not correspond to the expectations of investors at the time the decision to invest is taken. After all, what matters is the opportunity cost of capital, which is only available and is only relevant at the time the investment decision is taken. This, in turn, requires that current market conditions be taken into account.

In this sense, it is considered appropriate to include ex-post and ex-ante data with a weighting factor associated with each. PwC recommends the DMS and LBS databases (2008) for the ex-post approach and the studies of the Competition Commission (2008) and Welch (2007), for the ex-ante approach. ICP-ANACOM considers that the Damodaran database should also be considered for the ex-post approach, and the database of Bloomberg for the ex-ante approach, given that it is the database of reference.

Given the recent financial crisis, it is important that the risk premium already incorporates the impact seen as of the last quarter of 2008. To this end, according to PwC, a recent review conducted in a study carried out by Grabowski 4 should be taken into account.

It is likewise noted that consideration should be given to recent regulatory precedent (2008), identified by PwC, since this constitutes a good basis for comparison.

Given the above, consideration is given to the proposal set out in PwC’s report to consider the four sources of information cited to calculate the risk premium (ex-post data, ex-ante data, recent revisions and regulatory precedent), providing a degree of consistency, as appropriate and adjusted to the parameter that is to be determined. PwC used a weighting based on professional judgement for each of the sources of information, presenting two distinct scenarios in this respect. However, the view is taken that that there are no objective reasons to assign different weightings to these information sources. Therefore an equal weighting will be given to each source, corresponding to an individual weighting of 25%.

Geometric average versus Arithmetic average

Every day, investors make decisions to invest with respect to the purchase and sale of shares. There is no clear and binding commitment to maintain a share portfolio indefinitely without there being at least an implied intention to act on the decision to invest, even though this may not be carried through.

According to available data, it is not possible to clearly assess the relevance of all the shareholders who take a long-term strategy. Furthermore, ICP-ANACOM considers that it is unreasonable to assume, for purposes of ascertaining the rate of cost of capital, that an investor will hold their participation on an unconditional basis (i.e. regardless of any developments that may impact the value of the company) and for an indefinite period. In this respect, and so as not to underestimate the risk premium associated with the average PTC investor, this should be calculated based on the arithmetic average of the observations.

As a result of the methods set out above, the determination of the risk premium results from the average of results obtained, using the arithmetic average of the four sources of information (ex-post data, ex-ante data, recent revisions and regulatory precedent), (See Table 9).

Table 9 - Calculation of risk premium

Source

Weighting

Risk premium

Ex-Post

DMS and LBS (2008) - France

 

6.20%

DMS and LBS (2008) - Italy

 

7.70%

DMS and LBS (2008) - Ireland

 

5.10%

DMS and LBS (2008) - Netherlands

 

6.10%

DMS and LBS (2008) - Norway

 

5.60%

DMS and LBS (2008) - Spain

 

4.60%

DMS and LBS (2008) – United Kingdom

 

5.40%

Damodaran (2009)

 

6.50%

Ex-post total

25%

1.48%

Ex-Ante

Competition Commission (2008)

 

4.40%

Welch (2007)

 

5.75%

Bloomberg (2008)

 

6.45%

Ex-ante total

25%

1.38%

Recent Revisions

Grabowski (2009)

 

6.00%

Total Recent Revisions

25%

1.50%

Regulatory Precedents

CMT Spain (2008)

 

7.00%

ComReg Ireland (2008)

 

6.00%

Arcep France (2008)

 

5.00%

Total Regulatory Precedents

25%

1.50%

Risk premium

100%

5.86%

Source: PwC 2009 Report and ICP-ANACOM calculations
 
In this light, ICP-ANACOM considers that in the three-year period 2009-2011, the risk premium to be considered is 5.86%.

Notes
nt_title
 
1 Damodaran, Aswath, "Equity Risk Premiums", Stern School of Business.
2 Belgium (IBPT) - 35 years; France (Arcep) - 26 years; Norway (NPT) - 105 years and United Kingdom (Ofcom) - 107 years.
3 See PwC Report 2009 - pp. 31 and 32
4 Grabowski, J, Roger (2009), "Problems with cost of capital estimation in the current environment-update".