SIAS SpA – Deutsche Bank “buy” recommendation and risk free rates

Normally I tend to ignore any sell side ratings for the stocks I am interested in.

However, this time with the Deutsche Bank “buy recommendation” for portfolio Stock SIAS (target 7,70 EUR) I find it interesting how they justify their valuation in the summary:

Our target price of E7.7 is based on an SoTP, which values each concession with an individual DCF based on an 8.5% WACC (5.5% risk-free, 5.0% risk premium and beta of 1.3x). We subtract net debt and provision and we then apply a 20% discount to reflect lower liquidity than peers and risk of value destructive M&A. Sias trades at a 30% discount to the European peers: 2012E EV/EBITDA is 4.5 (vs. 7x for the sector) and 2012E PE is 7.4x (10x). Downside risks are regulatory changes, more negative traffic performances, capex delays and value-destructive M&A (see page 32).

This is highly interesting. As I have written in an earlier post about risk free rates, the CAPM requires to use the default free risk free rate in the currency of the issuer.

It is difficult to determine an “undisturbed” risk free rate in the EURO anyway and maybe the 10 Year rate for Bunds at 1,46% is too low, but using 5.5% as a EUR risk free rate is defintiely wrong. Even more as the Bonds outstanding from SIAS yield 4.9 and 5.4%, which is below the assumed risk free rate.

In my opinion, the Deutsche Bank assumptions double counts the Italian risk because they use a high Beta and a “risky” rate to discount rate. Just as another interesting point: SIAS beta relative to the Italian FTSE MIB is only 0.8.

Nevertheless it is interesting, that even with double counting Italian risk they still end up with a price target of 7,70 EUR which would imply a 65% upside from the current 4.70 EUR.

This shows how undervalued some of the PIIGS shares are at the moment.


  • Martin,

    you get all the info on SIAS homepage. Traffic declined in Q1 (some of the decline was due to weather and strikes), but increased tarifs comensated for that.

    That’s the nice thing with toll roads: “demand” is relatively inelastic to price as there are no real alternatives (i.e. = natural moat).


  • You are right. This investment looks promising anyway. For me I discount 12% for taxation of dividends (german perspective) + 8% corporate governance in Italy = 20%. The interest rates used by Deutsche seems too high, as there is no 5.5% risk-free in the euro-zone.

    Did they use historical DCF or estimates? I think one should focus on how realistic the future cash flows are. How much revenues does come from corporate and private customers? Corporate customers should decline as the economy does not look so good. E.g. look for import/export activities in Italian harbours. Eurokai reported declines 1st quarter 2012 for Italy. On the other hand private savings are quite high in Italy. So private customers should still be able to pay. Just my opinion.

  • What about taxation of Italian dividends? It is much more problematic to get this money back from the Italian state than from Swiss for example. E.g. Germany does only account for up to 15% of this tax, if you reside within Germany and get foreign dividends.

    • Martin,

      this tax disadvantage does not really change the investment case. What will be the impact over a 3-5 year holding period ? 50 Cent or 10% including the special dividend ?

      I can live with that,


  • interestingly, I have seen weird dcf assumptions from DB several times. In fact, why do they use a 20% lump-sum cut for lower liquidity and potential merger risk? It seems that the analyst’s MD was walking buy and saying “we don’t want strong buy rating on italian stocks these days”.

  • in theory, the country risk should be included in the volatility of the respective stock index or stock.

    The Italian market for instance shows historic volatility (50 days) of 31% aganinst 23% in the Dax.

    Using a beta of 1.3 for Sias AND a 5,5% “risk free” rate is double counting.

    I am not sure how they derive the Beta of 1.3 as SIAS has only a Beta of 0.8 against the Ftse MIB, but you should not do the risk adjustment twice.

  • “- you don’t add country riskk premiums to risk free rates in the EUR. The Damodaran table is to be used if Italy for instance issues an USD bond.”

    – how do you incorporate country risk then? In using high beta that measures your risk against the “average european stock” ? A stock could then be riskier for many reasons, country risk one of them?

    • Welju,

      i think we are arguing here in circles.

      My thesis is that Deutsch Bank (and maybe many others) are including Italian country risk multiple times into their discount rates. Of course you can use any rate you like but it should be consistent.

      As jan pointed out, the whole exercise gets even more useless when they do an additional 20% haircut.

      Of course I could be wrong and you have to discount Italian toll roads with 15% or more (although the corporate bond EUR ond of the same issuer yields less than 5%), but I am betting against this.


  • #welju,

    maybe i didn’t make my point clear:

    – the CAPM only works with current rates not averages
    – you don’t add country riskk premiums to risk free rates in the EUR. The Damodaran table is to be used if Italy for instance issues an USD bond.


    • Another example where Damodaran is using different country risk premiums in the euro zone: (p. 22)

      And he also said: “There are surprisingly large differences in the actual premiums we observe being used in practice”.

      • Welju,

        in the example you are referencing, he is referring to “risk premium”. This is the equity premium. In his example the risk free rate (for Nestle) is simply the Swiss Govie rate.

        It is correct to incorporate the country risk into the risk premium and/or beta, but not twice !!!


      • I see what you mean. But some (DB) people prefer as risk free rate for SIAS SpA the yield on Italian 10-year bonds and the 5.0% risk premium is probably the (historical) premium for the equity market. And talking about their beta, here I have no data.

        As said before: The risk-free rate is a soft variable.
        German 10-Year Bunds are not really risk-free.

  • 5.0 or 5.5% is probably the historical risk premium for Italy.

    You could also use bond ratings and appropriate default spreads: (Damodaran)

    The risk-free rate is a soft variable.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s