Summary
- At first glance, CSX is a slow-growing business that underperforms the market this year.
- With its strong capital allocation policy and significant cash flows, CSX could outpace the market in the long term, even against the backdrop of a potentially challenged growth environment.
- CSX has been able to consistently raise dividends and buy back shares – allowing for more dividend raises – on the basis of continually improving operating margins.
- Based on my dividend growth valuation model, CSX looks moderately undervalued.
Investment thesis
CSX (NASDAQ:CSX) displays little in terms of top-line growth but has demonstrated an ability to increase efficiency in operations resulting in improved profit and cash flow margins. With the increased profitability, CSX has delivered considerable cash returns to shareholders. While the industry faces challenges from the implications of a recent derailment accident in Ohio, US – which I will be discussing later – and a slowing economy, my valuation analysis based on CSX’s history of returning cash to shareholders suggests undervaluation to some extent.
Business overview
CSX Corporation is a US-based holding company with its main operating entity being CSX Transportation. CSX Transportation is a freight railroad company and one of just seven ‘Class I’ railroads – a term designated the largest US railroads measured against certain revenue thresholds. With its main competitor being Norfolk Southern Corporation (NSC), CSX and Norfolk Southern operate a duopoly-like rail freight business in the eastern United States.
CSX divides its revenue in four lines of business: (1) merchandise, (2) intermodal, (3) coal and (4) trucking. With $14.9 billion generated from its services in 2022, revenue was split between the four lines of business as follows:
Financials: CSX’s most recent quarter
On 25 January 2023, CSX reported its latest quarterly earnings (full-year 2022 results), slightly beating consensus estimates in terms of revenue and EPS. Diluted EPS demonstrated a 16% increase from the previous year at $1.95 per share against $1.68 for 2021. This means EPS growth was slightly slower than that of the top line with revenue increasing 19% to $14.9 billion.
On 15 February 2023, the company filed its 10-K. Following a slowdown and challenged growth during COVID, CSX has managed to grow revenues which came in at $14.9 billion for 2022. Operating income was $6 billion, displaying a 7.7% positive development from 2021.
In its 10-K, CSX summed up its results as follows:
If we zoom out and look at top-line development the past several years, it would suggest that CSX is struggling to grow, with revenue being more or less stagnant for the past 10 years.
The difficulty in growing the top line could suggest a struggling business. However, during the same period, earnings have more than doubled – and so has free cash flow. Both profit margins and cash flow margins have improved substantially through the period. This, on the other hand, suggests that while management hasn’t chased (or achieved) much in terms of top-line growth, the business has been run with an emphasis on efficiency instead of chasing top-line growth. Seemingly, the improvement in margins have expanded cash flows, and that cash flow is available for distributions directly to shareholders. As a matter of fact, as I will discuss later, management has used the improvements in earnings and cash flow for direct returns to shareholders.
Before moving on to those shareholder returns, I would like to draw attention to a key metric from the 10-K that underlines the point of operational efficiency just discussed. This metric is that of the operating margin. It measures the company’s operating expenses as a percentage of revenue and is key to analyzing the cost efficiency of a railroad compared to its competitors. For CSX, the FY2022 operating ratio was 59.5% – having increased 420 basis points from 55.3% the year prior. For 2022, CSX has performed better in terms of the operating ratio compared to the three other US-based Class I railroads as illustrated below (the lower the operating ratio, the better):
With CSX having the lowest operating ratio, it is essentially running a more cost-effective business – at least for the past year – than its competitors. While cost efficiency is always important in business, it is particularly important to the railroad business. This is because all railroads essentially do the same – they move goods from A to B using locomotives and freight cars. While technology, individual workforce skill and management are all important factors to performance, in running what could be described as a ‘simple’ business, cost efficiency becomes a top priority.
Whether or not CSX can maintain its lead in terms of cost efficiency obviously remains to be seen. Management – not surprisingly – seems keen to keep the position with operational efficiency being a key part of the company’s current ‘ONE CSX’ corporate strategy.
Shareholder returns
As noted previously, CSX has demonstrated an ability to increase cash flows. CSX has also demonstrated a willingness to return the cash to shareholders.
During 2022, CSX repurchased $4.7 billion shares and reduced its share count from approximately 2.19 billion shares to 2.06 billion shares, constituting a ‘buyback yield’ of about 5.5% for the year. The past year’s buybacks continue a long policy to return cash to shareholders through buybacks. CSX has consistently reduced the share count for more than a decade – even through the pandemic.
By the beginning of 2013 – just a decade ago – CSX had 3.06 billion shares outstanding (adjusted for stock splits). As per the most recent 10-K, that number has come down to 2.06 billion – constituting a drop in shares outstanding at about one third in 10 years.
Based on having declared a dividend of $0.11 for the past quarter, CSX’s full-year dividends should come in at $0.44 per share. This puts the dividend yield at approximately 1.5%. CSX has raised its nominal dividend every year for the past several years. Measuring dividends from the start of 2013, the track record complements that of the buyback history.
In order to assess the sustainability of the shareholder returns, attention should be paid to the company’s free cash flow and how the free cash flow develops relative to the shareholder returns.
For the full year 2022, CSX generated approximately $3.7 billion of free cash flow – money that is available for allocation to shareholders. Of that, $852 million was used for paying dividends, and $4.7 billion for buybacks. This puts the total cash returns for the year at more than $5.5 billion.
Between 2013 and 2022, CSX has gone from producing $895 million to about $3.7 billion of free cash flow. The below graph shows the development of free cash flow versus cash shareholder returns for the period:
As is evident from the graph, CSX has almost consistently through the period returned more cash to investors than it has produced (free cash flow). This suggests CSX has been borrowing money to maintain its cash distributions. This practice is often criticized – although it may make sense to utilize debt to buy back shares, particularly if low interest rates are obtainable which could’ve been the case through the years in question. Further, as the graph shows, far more cash is returned through buybacks than dividends. When shares are bought back, the dividend payments that would otherwise accrue on the shares bought back are voided, and the company ‘saves’ that cash. This allows for further focus on increasing cash returns through dividends – and as noted above, dividends have been consistently increasing through the period, whereas buybacks have been ‘rollercoasting’ through the period. Further – debt for buybacks aside – it is evident that CSX has generally been able to increase its free cash flow considerably through the period.
Valuation
For the purposes of this analysis, I’ll use the dividend discount model (DDM) to assess the per-share value of CSX. Three variables are included in this valuation: The current dividend yield, the expected growth rate (CAGR), and the required rate of return (for which I will use the weighted average cost of capital (WACC) applicable to CSX). Because of the inherent limitations to the simple DDM formula (which I will describe along the way), I will be providing a bull, bear and base case scenario to shed light on the possible value range.
‘Bull case’ valuation and commentary
For the bull valuation I will be applying the DDM in its simplest form using the metrics described above.
On 14 February 2023, CSX declared a dividend for the quarter of $0.11 per share. Assuming the quarterly dividend remains at this level for the entirety of 2023, the annual dividend comes in at $0.44 per share.
CSX has raised its dividend by an average 8.5% for the past 5 years. For the purposes of this valuation, I’ll assume an equal forward-looking growth rate of 8.5%.
The weighted average cost of capital (WACC) for CSX is 9.37%.
With the above input, each share of CSX – using the dividend discount model outright – is estimated to be worth just above $50. At its current price level just under $30, CSX appears significantly undervalued. The challenge with applying the DDM in its simplest form as done here is that it is assuming a perpetual stream of dividends (‘eternal’ dividend growth at a specific rate), and even small deviations from the growth rate can cause substantial variations to assumed value. Therefore, I will be walking through a modified valuation assumption based on potential (or likely) deviations in the dividend growth rate in particular – labelled here as a ‘bear case’ scenario.
‘Bear case’ valuation and commentary
In attempting to measure the effects of future deviations in the dividend growth rate, I will apply the ‘H-Model‘ which is a modification to basic DDM formulas. The H-Model accounts for the (inevitable) change in dividend growth rate by assuming a decline in the dividend growth over time:
Where:
The first part of the equation calculates the value of the stock based on the long-term (low) growth rate. The second part considers the value of the stock based on the initial high rate period – and the values are then added together:
Hence, the H-Formula accounts for deviations in future dividends by assuming that the initial (current) dividend growth rate will only be sustainable for a while, and after a while – the ‘half-life’ – it will drop down to a sustainable (lower) level and thereby modelling a more realistic picture of the value than the basic DDM formula.
Just as there are limitations to basic DDM formulas, one caveat with the H-Model is that it requires more assumptions than basic formulas. This again means that the modelled valued relies on subjective estimates. This is worth keeping in mind when comparing the modelled values from the applied formulas.
For the purposes of this valuation, I will assume that the 8.5% dividend growth used in the basic formula will be the ‘initial short-term high dividend growth rate’. Or in the other words the current but not sustainable growth rate (in the long term). I will assume that the company is able to maintain this rate for 12 years, meaning after that I assume that dividends will decrease to a lower growth rate. This lower growth rate – the ‘sustainable long-term dividend growth rate’ – I will assume is 6.5%, a number I believe resembles a mature ‘cash cow’ type company, at least considering the effects that buybacks can have on the ability to raise dividends in the long run. For the required rate of return I will use WACC as in the case of the basic DDM.
With the above assumptions, the modelled value per-share is just under $30. So under these assumptions, CSX appears more or less fairly valued by the market.
‘Base valuation’ and commentary
To ‘challenge’ my calculations in connection with the bullish/bearish valuations according to the DDM models above, I would like to analyze valuation multiples and compare them to past levels. CSX’ last 5-year average P/E ratio (fiscal) is 17.4. This compares to a current P/E ratio of 15.17. Assuming a value in accordance with the 5-year average, each stock should trade at about $34, suggesting undervaluation to today’s stock price, albeit to a modest degree. The suggestion of undervaluation is supported by the fact that CSX trades at a P/E comparable to that of 2014 (when the dividend was considerably lower, several more shares outstanding, and an overall lower cash return as demonstrated in the analysis of cash returns above):
In conclusion, the valuation methods used here suggest a ‘valuation range’ between $30 and $50+ per share. With the stock trading below $30, it would suggest the stock is undervalued at least to some extent and potentially to a substantial extent – and depending on the ability of the company to keep returning cash to shareholders which could push valuation higher (higher dividends made possible by buyback and cost efficiency means higher valuation).
Risks to consider
With the US pushing for greener energy and less dependence on fossil fuels, the ecosystem surrounding fossil fuel production may hurt in the long term. CSX is part of the fossil fuel ecosystem in that it transports coal from mines in the eastern US to various electricity utilities. As noted in the Business Overview paragraph, 16% of CSX’s revenue comes from transporting coal. This means one of the most important streams of income for CSX. On one hand, this provides a steady business, people need electricity and consume it every day, hence keeping up demand for coal and demand for it to be transported. You may think that potential competitors may opt to transport coal via truck. But as a matter of fact, transporting bulk loads – such as coal – via train is far superior from a fuel efficiency point of view, and therefore also from an economic standpoint. One freight train can carry thousands of tons of coal. Something that would take dozens of trucks (and therefore dozens of drivers etc.). On the other hand, relying on one ‘out of favour’ industry the way CSX does could mean that revenue (and income) growth will have to battle the downward ‘gravity’ from declining coal transportation.
A second – and very current – risk is the implications of the Norfolk Southern derailment accident in East Palestine, Ohio. While CSX was not directly involved in the accident as it involved a train operated by competitor Norfolk Southern, the accident has led to widespread public attention, which again could cause changes to the industry – be it from legislators or the carriers themselves. The accident is still under investigation, but one possible impact on CSX from the accident is increased safety requirements and the costs associated with that. For instance, it has been argued that better brakes on trains and train cars are necessary. This could be very expensive and time-consuming (in terms of maintenance required) to retrofit current train stock, which would challenge CSX’s low-cost operating efficiency. Others have argued that the industry prioritizes buybacks over safety, and that while billions of dollars have been returned to shareholders via such buybacks, safety is said to have been underprioritized in more general terms. While as an investor you would unquestionably demand strong prioritization of safety for both employees and the areas in which the company operates, I would challenge the notion that the railroad industry has ‘prioritized buybacks over safety’. The railroad industry spends billions of dollars on safety – and aside from adverse human impacts of railroad disasters, bad safety is also really bad business. To me, saying that ‘buybacks have been prioritized over safety’ is the equivalent of saying that wages for staff have been prioritized over safety. The fact is that running any business – and large businesses like railroads in particular – is like running an ecosystem of several relevant stakeholders. No safety means no railroad. No employees means no railroad. But – importantly too – no investors mean no railroad. The job of management is to make sure all stakeholders are taken care of and balanced. For the potential investor in railroads – CSX included – the point to be made here is that investing in CSX means accepting the risk that something will change to the adverse in the industry because of the negative attention the accident has caused. Be that through reduced buybacks in response to higher safety costs (which could hurt valuations), or in some other way.
I would like to touch upon a final risk factor here that I believe is important to keep in mind: The railroad business is very capital intensive. In total, CSX employs almost $42 billion of assets. Most of that asset base consists of real estate and other ‘real assets’ – locomotives, freight cars etc. The risk that this imposes is that running the business will always be expensive from a nominal dollar perspective – whether freight rates (the charge for using the rail services) increase or decrease, there’s always going to be large capital costs to maintain the operation of CSX’s large rail system. On the other hand, this capital intensity also is a hedge against new competition – a barrier to entry, if you will.
Conclusion: Final thoughts
In conclusion, CSX has a great capital allocation policy (buybacks and dividends) that I regard as a strong positive. The ability that CSX has demonstrated to operate a low-cost rail business measured against competitors – and the upward effect that has had on earnings and profit margins – boosted those cash returns. Coupled with an upward trend in free cash flow, it would also suggest that CSX is still on track to deliver shareholder returns along the same lines. Going forward, I would watch out for the risks associated with possible industry changes from the Ohio derailment accident (in the near term), and in the long term, I would follow the free cash flow generation of the company to assess whether the company can sustain the capital allocation policy that I emphasize.
Besides, the traditional distinction between buying long and short may not be very telling here. As a matter of fact – in terms of stocks like CSX – I would distinguish between being long and being very long the stock. With slow growth and a dependency on the year-to-year cash returns from dividends and buybacks to generate investment returns, it might take a long time to get a satisfactory return from CSX. The valuation models used here show a ‘value range’ that suggests at least modest undervaluation, but potentially substantial undervaluation – but it is based on a continuous holding period, and increases in dividends that seem likely only if buybacks are also continued.
For the reasons mentioned above, I rate CSX stock a Buy.