AENA:SM is the Spanish airport system. It is regulated on a dual-till structure, which means that aviation services are regulated on a Return on Capital Employed structure (not unlike U.S. utilities) but other services (retail, car park, on-site buildings, etc.) are not regulated.

 

As background, it is a mistake to simply line up infrastructure assets (or even airports) on an EV/EBITDA basis. There are at least the following differences among them:

 

- Tax rates

- Breakdown between regulated vs un-regulated

- Competitive Risk (esp 2nd tier int'l hub airports)

- Regulatory structure and transparency

- Acquisition risk as many groups looking to acquire smaller airports in less developed countries

- Project management risk if undertaking large new projects

 

Spanish tax rates are 25%. The non-regulated portion of AENA produces 50% more cash flow than the regulated portion and growing fast. AENA:SM has limited competitive risk as Spanish airports are not large international hubs. Any international hub traffic at Madrid or Barcelona would be upside. Its financials and price-setting regulation is transparent. The vast majority of the traffic is coming from or staying in Iberia. Its regulatory structure is clear and transparent. Its management has limited foreign investments. It has large excess capacity and so there are no large projects on the horizon.

 

Compare with the Paris Airports Group (ADP). French tax rates are 34%. ADP also has little competitive risk but for the opposite reason: they are the dominant top-tier hub of Western Europe and not likely to lose that position. The regulatory structure is similar to that of AENA, but ADP is not transparent between regulated businesses and non-regulated businesses in their GAAP filings. It is difficult to reconcile their price regulatory filings with their GAAP filings. Further, ADP management has made no secret of its desire to acquire stakes in airports outside France, esp in the Middle East. And last but not least, ADP is proposing (and interested parties are encouraging) a massive regulated capex spend over the next 15-20 years. The current proposal will double the regulated asset base in the next seven years at ADP and about quadruple the regulated asset base over the next 20 years..

 

If ADP management successfully completes its ambitious plans over the next decade, it is much cheaper than AENA:SM, but the risks are many, and it should not be confused for the safe "fixed-income" like cash generator that is AENA:SM.

 

Financials

 

Market Cap: 150mm shrs * E165 = E24,750mm

 

Net Debt is E6.9bn, but this includes E300mm consolidated from 51% interest in Luton airport. So net debt at Spanish Airports is E6.6bn, and the average interest is 1.25%. 

 

The three main non-Spanish airport assets are worth E1.0bn:

Lutron 51% equity: $0.4bn

Mexican GAP equity: $0.3bn

Spanish RE Services: $0.3bn

 

So we now have adjusted mkt cap of E23.750bn

 

In 2019,

 

Spanish Regulated EBIT will be E970mm. Unlevered FCF is then E728mm.

Spanish Unregulated EBITDA will be E1,040mm. D&A is about E50mm. And Unlevered FCF = E775mm.

 

Take out E62mm for after tax interest, and FCF-to-Equity is E728mm +775mm - 62mm = E1441mm. (or 16.5x for a very stable business growing at about 5% with little additional capital requirements).

 

However, this is a bit too good to be true, and this is where things get a bit fuzzy.

 

We need to normalize our numbers to take into account a few items:

 

1) They are over-earning on the regulated business due to passenger numbers exceeding expectations when rates were set in 2017.

2) Interest rates have dropped and this will also hurt unlevered regulatory earnings, but this is more than offset by lower rates on own debt.

 

There are two basic ways to think about the regulatory structure here. You can read the laws directly and learn that the future earnings are to match a cost-of-capital (calculated via a strict CAPM framework) multiplied by the RAB. In this case, if we reset the rates today, regulated EBIT would drop from E968mm to about E692mm. So they are overearning by about E276mm (or E207mm after taxes).

 

If we adjust for this, FCFE = E1,234mm. But they will over-earn by about E500mm until the next reset so subtract this from Adj Mkt Cap. We also need to adjust for lower rates. And we arrive at about 18.25x FCF-to-Equity (or about 5.5% yield).

 

The second way to look at regulatory structures is that the allowed cost-of-capital is backed into via a negotiation, and that what regulators are most concerned with is consistency in fees per passenger. In other words, if passenger numbers were to plummet, regulators are not going allow fees to rise accordingly. But this is also true in reverse. and the good news is that AENA is over-earning and so this should work in their favor. I do not expect regulators to reset rates according to strict formulas and they will smooth the fees per pax. In other words, if the reset period were now, I would expect a new regulated EBIT between E692mm and E968mm and closer to the higher number.

 

Rambling Aside: Applying the strict formulas is virtually impossible because calculating the forward equity market premium and the beta of the regulated assets is guesswork. For example, the beta of regulated assets, if reset properly, should be tiny. They can under-earn or over-earn during short periods until next reset, but long-term beta should be tiny. However, we can look across airport regulators and the cost-of-capital is much greater than a 0.1-0.2 beta would allow. And in real life, these regulated-asset betas are not observable. I have had many conversations with airport management and even a couple regulators over the past decade, and they are beyond vague about how these numbers are derived. There are three categories: (1) the most vague will not reveal anything other than fees per pax; (2) the middle group adds the cost-of-capital; and the (3) the final group reveals beta and equity mkt premium. But I have never had anyone tell me how the beta and mkt premium are derived except in the vaguest terms by referring to academic CAPM model, which only begs the question.

 

The Paris Airports are a year ahead of the Spanish airports in their 5-year regulatory window and have made their initial public proposal for 2021-2025. They have the cost-of-capital increasing slightly over the terms set for 2016-2020. This is only a proposal, but this proposal is not coming from a vacuum. Management has consulted the relevant parties, including regulators, and the proposal supports the second way of looking at regulatory structure.

 

So why don't regulators slash the cost-of-capital and fees per passenger? Several reasons: 1) to do so would indict the past process, not something regulators want to do; 2) the equity is 51% owned by government, so it is a nice revenue generator; 3) flying is a voluntary activity engaged in by above-average-wealth people so not a populist issue; and 4) local and federal governments have projects they want completed and this gives mgmt leverage.

 

One other aside: The capex here is slightly lower than than the D&A and so RAB is dropping about 1.5% per year. It is important to realize that regulated EBIT will also drop 1.5% per year according to formula, but that capital is returned to AENA. I run the numbers using capex = D&A for simplicity. You can fire up the spreadsheet and see that value not much different if allow RAB to drop 1.5% per year. What you cannot do is use lower actual capex and then assume regulated EBIT stays constant. This might happen in reality (see second framework above), but certainly contradicts stated regulation.

 

---

 

The unregulated business has an extremely high ROIC. It has about E1.3bn of PP&E and produces about 750mm of unlevered FCF. Its ROIC is about 50%. This business grows with # of pax and amount of spending per pax. Both are related to economy. We have a fair amount of data related to GDP; passengers; and non-regulated EBITDA.

 

The basic relationship between passengers and real GDP is pax growth is about 2x real GDP growth +1%. On the downside from 2008 thru 2013 (recessionary the entire time), pax dropped about 11% while GDP real dropped about 9.5% peak to trough. So on the downside, pax growth was less levered to real GDP growth, which is nice.

 

Revenue and EBITDA are related 1:1 to number of passengers (plus inflation). The business has very little operational leverage as the EBITDA margin is almost 80%. The relationship has been fairly precise as these things go: EBITDA growth = inflation + 2*real GDP growth + 2%.

 

As a check, from 2014 to 2019, unregulated EBITDA growth per annum has exceeded 10% while real GDP growth has averaged 2.8% and Spanish inflation has averaged about 1%. During this period, the European economy has been listless and the Brexit saga has affected traffic from the UK.

 

Over the next decade, I expect Spanish real GDP to be between 1-2.5% per annum and inflation to be between 0% and 1.5%. Let's take the low end of both, and we would expect about 4% EBITDA growth per year in the unregulated business. This would translate to about 3% growth in the total business. And clearly, this business's profit is much less levered to economy than average non-financial company.

 

So even with fairly pessimistic economic and regulatory assumptions, we would expect a FCFE yield of 5.5% growing 3% per annum for an 8.5% IRR. And in mid case, I expect 6% unregulated growth for a 10.0% IRR.

 

And in the event of a major recession where Spanish GDP falls 5%, I'd expect unregulated EBITDA to drop HSD based on 2008-2013 recession. Hardly a disaster.

 

Aside: Airport retail value runs counter to traditional retail value. As malls and urban retail value plummets, brands increasingly need to get their products in front of physical eyeballs. And airports, with their captive, relatively wealthy audience may be the last remaining venue for this. In this way, airports should also be considered as media rather than traditional retail space. And this trend will only continue as traditional retail spirals downward.

 

Risk: The major risk here is that I have under-estimated the chances of a significant change in regulatory inputs and a slashing of the regulated EBIT.