Estimating earnings is really a case-by-case process. There is no fixed procedure or static set of metrics one can look at to accurately forecast earnings. I should rephrase that. Of course there are metrics like sales, margins and balance sheet numbers one can look at, however different companies present different stories even when these companies are very similar. Since they present different stories one must use different methods. It is indeed a much nuanced process. For example, while Verizon is struggling with higher wage and benefit costs as a result of a lost battle with unions, CenturyLink (NYSE:CTL) is comfortable firing thousands of employees. Unfortunately this doesn't seem to be enough to stop the margin compression the company is facing. When looking at the financials we can see an inverse relationship between the cost of revenue and the revenue itself, while the latter decreases the former increases. I believe that CenturyLink might very well miss their earnings estimate as a result of this trend. The good news is that the LVLT acquisition makes up for one underperforming quarter for two reasons:
1) CenturyLink is acquiring $10 billion worth in NOLs through the acquisition
2) The acquisition will provide a solid basis from which margins can stabilize.
Increasing cost of revenue while revenue declines
EPS in dollars
The first thing one notices when looking at CenturyLink is that analysts are expecting significant EPS growth when compared to the previous three quarters. To find out why, we must dig in to the balance sheet numbers and previous reports because we must understand the earnings drivers of this specific company. Only then can we zoom in on these factors, analyze them and determine whether re-adjustments are required.
When comparing quarters there are a couple of things that stand out. The cost of revenue has risen sequentially since 1Q16 while revenue levels decreased. This already is a clear sign of a business that is under pressure: spending more to sell less. As a side note, this is a development that long-term investors should watch closely as it reveals the strength of the business model. In CenturyLink's case, this is definitely a negative development which will dampen the company's (and thus shareholders) profits. From 1Q16 to 3Q16 the cost of revenue increased by $96 million or roughly 18 cents a share. It seems that we're in luck, because this roughly explains the drop in EPS from 0.44 1Q16 to 0.28 in 3Q16. This doesn't seem to be the whole story since CenturyLink reported an adjusted EPS of 0.56 in the third quarter. If all else remains equal, and the logic of adjustments are followed i.e. they will not re-occur in 4Q16, my work here is done. At least the part where I reconstruct the guidance and try to get in the head of Wall Street analysts. Of course, now we need to know why this is the case and also whether this is a trend or whether this represents short-term headwinds. According to CenturyLink and analysts these were short-term headwinds. So let's see if that is indeed the case.
The adjusted EPS is as a result of adding back "purchase accounting intangible" and amounted to 24 cents a share. The other 4 cents a share is what management calls special items which include severance costs, net loss associated with early retirement of debt and a tax effect relating to the fourth quarter of 2014. Of course the most interesting part is the 24 cents a share add back.
When looking at the numbers we can see that the bulk of the add back is related to adding back amortization of customer base intangibles which add $131 million (24 cents a share) to CenturyLink's net income after accounting for tax. I do not agree with this number, because amortizing intangible assets is standard practice and is done over a specific number of years regardless of the actual useful life of the asset. I see no reason as to why CenturyLink should readjust for amortization since this is common practice. This is indeed a non-GAAP practice, but what's even more interesting is the fact that the decrease in net income is perfectly explained by the higher cost of revenue. I highly doubt that GAAP earnings will come in at 56 cents a share in 4Q16. Before I get to my conclusion on what I think this means we must add the last piece of the puzzle.
Centurylink to increase EPS by cutting jobs
While Verizon was recently forced to hire more workers and pay its existing workers higher wages and higher benefits, CenturyLink is actually cutting jobs. Management wants to reduce the headcount by approximately 8% or by 3,400 and most of the layoffs are expected to have occurred by December 16th. This is a very good second clue as to why the 4Q16 guidance (0.56) is so much higher than the previously reported 3Q16 earnings (0.28). Of course, we can't just point to headcount reduction as our magical fix without running the numbers.
If we're being generous and count all S&GA costs as strictly employee salaries we can just reduce that number ($796 million) by 8% to arrive at a cost saving of $64 million or roughly 12 cents a share. Keep in mind that while S&GA expenses include overhead expenses like salaries the number is not limited to salaries exclusively and also includes other overhead expenses like rent/mortgage on buildings, utilities and insurance. The point I'm trying to make Is that the cost saving is likely less than 12 cents a share, but it is certainly no more than that. We also have to consider the fact that layoffs are often accompanied by severance packages so I expect there to be some form of expense as a result of the layoffs. Granted these expense will be one-time restructuring expenses. Even though it is reasonable to assume some level of cost, it is actually quite hard to get a grip on how much it is going to cost the company.
If we add 12 cents to the 3Q16 number we arrive at an EPS of 40 cents, which is still significantly lower than the expected 56 cents a share.
Looking forward: the acquisition
The second factor that might boost Centurylink's earnings is their recent acquisition of Level 3. The deal is set to close by the end of the third quarter of 2017 and would add roughly 140 million or 26 cents a share in net income to Centurylink's bottom line. I must admit that mergers are not my forte, but I do believe that the additional net income cannot be consolidated at this time since the deal hasn't closed yet and will take approximately eight months to close. The acquisition namely serves the purpose of acquiring LVLT's NOLs which are worth $10 billion. It also helps CenturyLink in another area. Somewhere in the start of the analysis I mentioned a concerning inverse relationship: revenues decreasing as costs of revenue increase, which means that the gross margin is declining. CenturyLink is attempting to solve this problem by acquiring Level 3. A more nuanced phrasing would be to state that CenturyLink is trying to counter act this development instead of solving the margin compression problem organically.
To see if this strategy might work we need to look at some of the numbers on the income statement of LVLT. With a revenue of roughly $8.1 billion LVLT's revenue is approximately half that of CenturyLink's. This is good news because this means that LVLT's gross margin can significantly improve CenturyLink's gross margin once the companies finally consolidate. So by how much will CenturyLink's margin improve? To answer this question we need to take CenturyLink's gross margin of 56.5% and LVLT's gross margin of 50.2% and create a weighted average by keeping in mind the difference in revenue size. The result of this calculation is a gross margin of 54.5%. This doesn't seem to be a positive development, but I would argue that it is because LVLT's gross margin enjoys a luxury that CenturyLink's gross margin doesn't: LVLT's gross margin is not steadily decreasing, but is actually remaining fairly stable. What this means is that as CenturyLink's gross margin compresses, LVLT will be there to break the fall.
Adding it all together
I'm not convinced that CenturyLink will report 56 cents a share in GAAP earnings, but I'd like to work out some different scenarios.
1) CenturyLink again readjusts for amortization of intangibles and indeed reports an adjust EPS of 56 cents a share. If this is the case CenturyLink will actually report around 68 cents a share in NON-GAAP earnings since they've realized a cost saving of approximately 12 cents a share. This would be seen as a huge earnings beat provided investors do not "catch on" to what is really happening.
2) CenturyLink reports 44 cents in GAAP earnings and misses the target significantly. It is not unreasonable to imagine that this would send the shares down.
3) CenturyLink reports both: 44 (0.28+.012+0.04) cents in GAAP earnings and 68 cents in non-GAAP earnings. This is the scenario that has me troubled for multiple reasons. The first reason is that this seems the most likely scenario and the second reason is that I have no idea on whether investors would indeed "catch on."
I seem to remember from my accounting experience that companies are required to report GAAP financials in the 10-K (annual report). So adjusted non-GAAP EPS shouldn't be a problem if one is looking for a bearish position. This is actually a very nice bearish earnings play if CenturyLink is indeed reporting 44 cents instead of 56 cents a share. The problem with this is that this rule isn't exactly enforced with vigor and thus diminishes my confidence.
Forecasting earnings is a tedious business and, depending on how much money and time one has, is sometimes not worth it. Because there is no quick model it can be a time consuming endeavor especially if the conclusion ends up being that there is no "play" here. As an earnings speculator I do not have enough conviction to construct a position around this thesis. I hope that this information is useful for medium- to long-term investors who are insecure about whether or not to hold through earnings. Whether to hold or not also depends on one's entry point, but if I had to make a decision I would hold through earnings if I believed in the business.
Even though my approach was centered on estimating upcoming earnings, it revealed a couple of things that I think long-term investors should keep in mind. The inverse relationship between cost of revenue and revenue is a big warning sign for me and should be a warning sign for other investors. CenturyLink is trying to counter act this by acquiring LVLT. As I have previously detailed, this seems like a sound strategy because LVLT's margin seems more sustainable. So even though I believe that CenturyLink investors will likely feel some pain in the upcoming quarter, they should rejoice in the fact that management is adequately fixing their problems. In turn this means that the generous 8.47% dividend will be safe for the time being.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
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.