On this article, I analyzed Danaos [DAC] as an ideal Private Equity investment as it brings a relatively stable cash flow profile together with a high leverage at a competitive cost (3-4% on average) which a perfect Private Equity example of how PE used to make money back in the 80’s.
In this article I will try to analyze how Danaos compares to its peers.
You can skip this part if you want to go directly to the core of the analysis but I think it is relevant to be able to check the details specially when making comparisons. These are the remarks that I think are more relevant of the numbers I’ve done:
- In market cap I am only including common shares. Preferred shares, especially relevant in ATCO and CMRE, are counted as debt. Therefore, they are included in the net debt amount and in the average cost of debt calculation. Despite preferred dividend is only mandatory if the company pays dividend to common shares, I have taken in account its nominal yield as cost of debt regardless of dividend payments.
- Capital leases are included as part of the debt, and therefore counted in the enterprise value of the company as well but they are assumed to have a 0% cost and therefore decrease average cost of debt.
- When the company is not transparent in the cost of debt of its different facilities, I have used a standard 4% cost.
- Enterprise value, or EV, is the addition of market cap, preferred shares market value, bank debt and capital leases net of cash and adjusted for non-business essential assets as cash, JVs, restricted cash and assets for sale.
- I deduct non-business assets with some haircuts. For JVs I apply a 60% cut in its balance sheet value, except for ATCO where I only applied a 20% haircut given how recent and relevant is the power business ATCO is invested in. I apply a 60% haircut in assets for sale and a 40% haircut for restricted cash. No haircut for cash.
- Normalized EBITDA figures are calculated based on past performance and expected revenue and margin estimates by consensus analysts. I tend to be a bit contrarian being less optimistic for very high growth rates or a bit more optimistic when growth rates are deeply negative. Other than that, is just the average of past and expected performance.
- Regarding operative cash flow it is a bit more difficult to be able to compare one to another as fleet ages are different and IMO extraordinary capex implementation degree is also diverse. Besides, not all companies provide the information in a comparable basis or in no basis at all.
- I have applied the same maintenance capex figure of 3.5% of revenues per year to all companies. This is important because actual capex figures will depend much more on specific fleet issues that I did not take in account.
- Normalized cash flow (OpCFn) is equivalent to normalized EBITDA but net of maintenance capex and taxes, if applicable. It does not deduct interest expenses nor principal repayments as I want to compare OpCFn with enterprise value. I want to know how much the business yields independently of how it is financed. That cash flow is not free cash flow nor free cash flow to equity holders.
There are many companies involved in container transport that do not do exactly the same thing as Danaos. There are liners that operate their fleets directly, fleet owners which own mixed fleets of container vessels with dry-bulks, tankers of other ship types. There are also companies that build or own containers. There are also companies that integrate marine, train and truck container transport.
We want to look at container fleet owners only. I have cut the comparable peers to the following ones:
Source: Company Filings and my analysis.
Sea Span – Atlas Corp
Sea Span is part of Atlas Corp (ATCO), a holding company with interests with recent exposure to power business. Container shipping remains nevertheless its main activity. ATCO It is the largest market cap company of the group and owner of the largest fleet with 125 vessels with more than 1 million TEUs in capacity which is more than two times Costamare, the next fleet owner by market cap.
ATCO is focused on larger vessels. Its average size vessel is 8,400 TEU, which is the highest average of the group. Its fleet is also the more modern of all, with only 7 years of average age, as the company has been expanding its fleet in the last years.
ATCO also went through its own restructuring phase in 2017-19 with new management (David Sokol), debt restructuring, and new capital (Fairfax). One aspect of the quality of the company’s fleet may be seen in the high debt per TEU that lenders have accepted as net debt adds up to c. $4,000 per TEU, also the highest level of the group.
The cost of debt of ATCO is nevertheless moderate (3.8%), and relatively comfortable in terms of scheduled amortizations. Its sources of capital are diversified: it counts with some preferred shares ($200M-8%), bonds ($600M-5.8%) and bank financing ($3.1B-3.1%).
Its revenues have consistently grown in the last five years, 10% on average and they are expected to grow even more in the next two years as a result if its fleet expansion. Besides, its growth has not been at the expense of margin, as its average EBITDA margin in the last 5 years is 68%, at the top of the group.
Its enterprise reaches $6.4B, equivalent to roughly 7x EBITDA or operative cash flow. That EV is also equivalent to paying c. $6,100 per TEU, the highest price per TEU of all the group.
Costamare (CMRE) is the second player by market cap and fleet capacity, with 482 thousand TEUs among a variety of ships spanning different sizes and ages from small feeder ships to ultra-large containership vessels (“ULCV”). Average fleet size per vessel is 7,000TEU and average fleet ages is 10.6 years.
The cost of debt of Costamare is among the highest of the peers. The company relies heavily in preferred shares, which cost an average 8.57% on $353M. Together with bank debt and capital leases, the total debt of the company adds up to $1.9B, which is equivalent to c. $3,500 per TEU.
Being tilted to preferred is costly but at the same time, as they do not have a maturity, it offers maximum flexibility of principal repayment. As it is the case with ATCO, CMRE does have a manageable debt schedule for the coming years.
The company revenues have remained stable in the last 5 years and are expected to grow by 10% in the coming 2 years due to its recent acquisitions and its orderbook of newbuilds.
Its average EBITDA margin in the last five years is 63%, lower than Sea Span but not significantly lower, despite the company has lower vessel operating expenses per day ($5,000 vs. $5,900.
The company EV is c. $2.5B equivalent to $5,100 per TEU or 6-7x normalized EBITDA/operative cash flow. Value per TEU is c. $1.000 lower but it is more or less in line in terms of value for $ of EBITDA or cash flow, implying that the for the same TEU, CMRE extracts more EBITDA or cash flow from its average TEU.
Danaos is the third company by market cap and by fleet by capacity, with 353 thousand TEUs. Its fleet is moderately old (11.4 years) and focused in mid to large vessels. Its average vessel size is 5,600 TEU which is significantly lower than ATCO or CMRE and quite in line with GSL.
I have already analyzed the company evolution in the last decade in this article, but let’s only mention that its debt has been fully restructured and presents a very competitive cost (3.7%) and a manageable maturity profile, at least for the coming three years. The management of the company is weak on the financial side, but banks are present in the board now, as they were forced to accept shares for debt in 2018, so there should be some pressure to pay back debt and to keep a conservative financial profile for a while.
Its revenues have been decreasing in the last five years due to fleet rationalization but is expected to remain flat for the next two years. Its EBITDA margin is at the top of the peer group, with 68% and it is expected to remain stable in coming years.
The company EV is $1.6B equivalent to c. $4.600 per TEU, lower than previous peers. The ratio of market cap to EV is 12% the lowest of the group, together with GSL. If we look at EV multiple to EBITDA or operative cash flow, it shows a slightly better multiples, cheaper price, than CMRE or ATCO, 5-7x instead of 6-8x. There is some difference, but it is not really significant.
Capital Product Partners
Capital Product Partners (CPLP) market cap goes down to $134M with a 100 thousand TEUs fleet. CPLP owns 14 vessels with an average age of 8.2 years, on the lower part of the group. The company is focused in mid-sized vessels as its largest ships are in the 10,000 TEU category while its average vessel size is 7,100 TEU, not that far from the top.
The company recently focused its activity on containerships after releasing the last vessel of its dry-bulk fleet. The fleet seems competitive in terms of vessel operative expenses per day, which is the lowest of the group.
The company leverage is similar in proportion to the previous companies and its average cost, 4%, is quite competitive and in the lower part of the range with comfortable schedule of repayments for at least three years.
The elimination of dry-bulk exposure makes sense given market conditions and explains why the company revenues have been decreasing 11% on average in the last five years and why it is expected to grow for the next two. Its EBITDA margin on the last five years has been 62% on average but it is expected grow in the next two years, probably also because of that strategy.
Enterprise value is $482M equivalent to c. $4,800 per TEU or 6-7x normalized EBITDA or operative cash flow, quite in line with previous peers.
Global Ship Lease
Global Ship Lease (GSL) is the fifth company with a market cap of only $128M. Its fleet is nevertheless 254 thousand TEUs, higher than CPLP and not that far from DAC. The average vessel, 5,900 TEU is similar to DAC, while its larger vessels, 9,200 TEU, are significantly smaller. Its fleet is, together with Euro Seas, the eldest of the group, reaching 13 years. Its vessel daily operating expenses are in line with ATCO and DAC which means the company is more competitive than them as its average size is smaller and its costs should be higher.
The company merged with Poseidon at the end of 2018 and attempted to restructure its debt in 2019 with no success. As a result, its debt is costly and not very flexible. Together with preferreds, total gross debt reaches $940M at an average cost of 6.35%. It has preferreds with no maturity paying 8.75% but also bonds, maturing in 2022 costing 9.88%. Banks bring average cost down but probably their maturity is short as banks will want to avoid going behind the bond term.
This structure is not that un-natural to shipping before the last restructurings that took place in 2017 and 2018. GSL restructuring has not taken place and it is a pending issue. It is not the level of debt that is worrying (debt per TEU is $3,500) but its cost and its maturity schedule.
The company performance in the last five years has been very good in terms of growth (+14%) and margin (65%) and is expected to remain positive for next two years.
Enterprise value is $980M, mostly comprised of debt, as market cap is only 13% of that EV, in line with the case of DAC. That EV is equivalent to c. $3,800 per TEU, the lowest of the group after the smaller Navios and EuroSeas. In relation with normalized EBITDA or operative cash flow EV multiples hover around 6x, very much in line with DAC, and marginally cheaper than ATCO or CMRE.
Navios Marine Container
Navios Maritime Containers (NASDAQ:NMCI) is the container play of Navios Maritime Holdings, which also owns other fleet types. NMCI owns 29 vessels with an average size of 4,900 TEU, most of them relatively small but also owning 4 larger 10,000 TEU vessels, with an average age of 12 years.
The strange thing in NMCI comes from its low EBITDA margin which has averaged 46% and reached a maximum of 50% in 2018. This probably has to do with its huge vessel daily operative expenses of c. $11,500, which are more than twice the rest of the companies in the group.
I read on some article that this was due to the management contract signed, doubtfully at arm’s length conditions, with the company manager which as it usually happens in shipping signed with the main shareholder and typically founder of the company. I have not dug deeper there but it seems a reasonable explanation. In any case this 15-20 percentage point difference puts NMCI in a different basis than the rest of the companies.
Its debt is not too high ($1,670 per TEU) and is not too costly (4.1%). EV reaches $280M, equivalent to c. $1.950 per TEU, which is the lowest level in the group but given its lower than average margin it does not make sense to look at EV per TEU.
EV to EBITDA or operative cash flow is in the 4-5x range, the lowest levels in the group. Smaller fleet and market cap and probably not the best governance image may explain this cheapness in terms of EBITDA.
Enterprise value per TEU seems like a logical valuation bar to compare one company to another. EV to EBITDA is a good comparison metric also. EV to net operative cash flow or EV to NOPAT, which is the same as net operative cash flow when maintenance capex equals depreciation, is, more than a comparison metric a good valuation metric.
The problem in the case of fleet owners is that those metrics do not take in account relevant business differences, notably the contracted time charter, or the backlog of each company.
ATCO, with $4.2B of contracted revenues to be received in the coming 4.2 years cannot be worth the same if that backlog was null and the company only operated in the spot market, yet the previous valuation metrics would not be affected by that difference.
Or when a company has committed to buy additional capacity its future debt will be higher and yet that future debt is not counted as part of the current enterprise value of the company.
In my opinion, those are the two factors that should be relevant for investors and that cannot be taken in account in the standard EV to operative net cash flow metrics.
That is why I have used and adjusted EV metric, that deducts from the standard EV a rough present value of the backlog of each company, adjusted by its EBITDA margin, and adds the orderbook off-balance sheet commitments.
The calculation for each company is shown in the following table:
Source: Company Filings and my analysis.
When adjusting for backlog and orderbook, the differences in valuation per TEU tend to be increased but they do not seem to change the picture completely, as we can see in the chart below.
If we take ATCO, the largest and more expensive company in terms of EV per TEU, as the comparison basis, we can see that the discounts to that reference increase when we take in account adjusted EV.
For example, CMRE shows a 20% discount to ATCO if we look at EV per TEU but that discount increases to 40% if we take AEV per TEU in account. In general, the companies in the group, except for Navios, trade with a 40-50% discount to ATCO in terms of AEV per TEU, instead of 20-40% discount if we look at EV per TEU.
So, adjusting for backlog and orderbook, the companies that looked cheaper, look even cheaper or, in other, words, the reasons for their relative pricing do not seem to be found in fundamental factors as backlog and orderbook but probably in market cap or fleet size.
It may also be an institutional investor bias. If you are a daring institutional investor betting on a risky sector as container shipping it is probable that you hedge your own decision by going for the largest company of the pack.
In the case of Navios NMCI, I think it is obvious that its lower margin and foggy governance do explain the difference. In the rest, I am not able to talk about a fundamental factor that explains it.
Which one to buy then?
AEV per TEU only tells us which company is cheaper in relative terms, but it does not tell us where we can make more money from the business cash flows. For that, we need to look at yields on EV and yields on market cap and how secured are those yields.
Net operative cash flow yield on EV, or the inverse of it, which is the multiple, is a standard measure that I use. As we can see on the table, the differences are not that big between the different companies:
Source: my own estimates on Normalized, revenues, ebitda margins and net operative cash flow (see methodology section at the start).
The cheapest EV, or highest yield, would be NMCI and GSL. If a foggy corporate governance is a no-go for you as it is for me, that leaves us with Global Ship Lease as the most obvious bet, followed by DAC and CPLP.
As there are significant differences in debt structure and cost between the companies, I think it is also useful to compare a post interest cash flow metric to market cap, as it is shown in the table:
Source: company fillings and my own estimates on OpCFn (see methodology section at the start).
With this adjustment, and leaving NMCI aside, DAC becomes the highest yielding bet for the investor, as the company has lower cost of debt than GSL.
Despite the last run of DAC share price, the operative cash flow, net of annual interest payments yield on market cap is an impressive and unbelievable 90% yield, or a 1.1x multiple, which implies that you need only a little more than one year of operative cash flow after debt cost to recover the investment in the company.
As I discussed in the Danaos article this is what happens when equity value is so depressed and the investor can buy a 12-15% yielding asset with only investing 10% the business market value, being the remaining 90% a reasonably priced and structured debt.
GSL is not far behind, with a 75% yield. CPLP comes next with 40% and then CMRE and ATCO with a very healthy 30% yield that is only low in relative terms to DAC or GSL but it is not low by any means when compared to other industries or sectors.
Besides, if you multiply that yield on market cap by the number of years of normalized revenues in backlog, then you get the idea of the total yield on investment that is covered by the current backlog.
DAC has a 90% yield on market cap and current backlog compared to normalized revenues implies that the company has 2.8 years of secured revenues, which implies a total of 249% yield (no compounding) on current share price simply from the current backlog. So, on the current $8.4 share price this rationale implies that the investor could make $20 per share in the coming three years from the recurring cash flow after debt cost payment (not principal).
After those three years, you would still own the company’s assets and liabilities and a running concern business.
GSL and CPLP would be the next more interesting options by that metric with a 166 and 143 percentage points of yield in backlog. Then with some distance the next option would be CMRE, with 114 percentage points and finally ATCO with 89%.
I think that the whole sector is attractive in pricing as the highest EV/EBITDA of the group is 7x of ATCO and 5-6x for the rest of companies, equivalent to a 12-15% yield.
EV combines net debt, at nominal values, and preferred shares and equity at market value. It is the amount that would cost an investor to buy the whole business which is equivalent to the whole assets cost for the investor, regardless of how they are financed.
EV yield in that regard is equivalent to the Return on Assets (ROA) for the investor. A business that offers a 12-15% ROA is a reasonably good business. It is not great, but it is certainly better than the average business, specially taking in account how capital intensive this business is.
Return on Assets is split between debt and equity and here is where the deep undervaluation of these companies’ equity comes in to play.
The highest price to book (PTB) is the one of ATCO and it is only 0.6x, but most companies are below 0.4x and DAC and NMCI are below 0.2x, which implies an 80% discount of market cap to equity book value.
Because of this, most of the total assets cost for the investor (enterprise value) are financed with debt, from 65 to 87%, depending on the company. The investor has only to deploy 15 to 35% of the whole assets cost.
The leverage that the investor gets, for example in Danaos, is close to 8 times, but the company is paying 3.7% on its whole debt. Its cost has been set in a company restructuring in 2018 based on the lenders-borrower negotiation taking in account loan to fleet value ratios, EBITDA and debt levels, etc…, but not based on how high or low equity book values were or are.
That is why the investor (not Danaos as a company) gets an 8x leverage without having to pay the cost that a leverage like that would require. That is a free lunch to me and is more than welcome.
The result of this, is that the yield on market cap is high to unbelievably-high depending on which company you look at. It is nevertheless important to notice that this yield is not to be compounded unless share price stays constant.
If share prices increase to reflect the cash that is generated for the equity investor, then the yield is not reinvested and therefore cannot be compounded. In any case a simple 30 to 90% simple yield for 3-5 years makes for a wonderful compounded return, as we can see on the table.
For me NMCI is not investable and Danaos is the next one on return potential. Then GSL and CPLP would be the following ones, but GSL’s debt is still to be restructured. CMRE and ATCO are the largest play, by market cap and by fleet size and probably offer more professional management teams than CPLP or DAC.
Disclosure: I am/we are long DAC. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Editor’s Note: This article covers one or more microcap stocks. Please be aware of the risks associated with these stocks.