IBM: 1Q'16 Financials Are Misleading

Summary

• IBM has a new way of segmenting its business for financial reporting purposes.
• The segments are bogus, hiding continued weakness in emerging businesses that are the future of the company.
• Earnings of old businesses are being used to prop up new businesses and this isn't disclosed.

This is my first post for Seeking Alpha. I mainly post on my own tech blog, but I've been straying too far toward the business side lately and my readers there are telling me to stop: they are tired of reading about IBM (NYSE:IBM). That's certainly not the case over here, so I'll begin with what I hope is a withering criticism of a certain type of post that occurs here far too often - trolling through public company financials and drawing conclusions without knowing much about what the company actually does. That's the case with many such posts here, written by wonks who really need to get out more often. So here's a short look at IBM's 1Q'16 financials from the perspective of someone who actually knows the company.

To set the scene a bit, IBM is a huge old company going through a very public transition from old products and services to new ones, the challenge being to build revenue from the new sources faster than the old sources decline. It hasn't been going especially well for 16 quarters now but still IBM can point to some significant progress here and there in its new technology areas - cloud, analytics, mobile, social, and security (called CAMSS). It is very important to current IBM management for these areas to grow and grow quickly.

IBM has recently changed how its businesses are functionally grouped and their earnings reported, which creates something of a challenge relating actual business activities to the strategic imperative areas defined by CAMSS. CAMSS is still what's important for the future, we're told, but the numbers work a little differently. The new subdivisions are Cognitive Solutions Segment, Global Business Services Segment, Technology Services and Cloud Platforms Segment, Systems Segment, and Global Financing Segment. For this post, I want to concentrate mainly on the Cognitive Solutions Segment.

Here I'll turn to the 1Q'16 earnings call transcript and the words of IBM CFO Martin Schroeter:

So now let's turn to the segment performance, and I'll start with Cognitive. I mentioned that Solutions Software includes analytics, security and social. To give you some context - analytics is the largest part of our Solutions portfolio, and includes our traditional analytics platform, industry solutions, commerce and the Watson related businesses. Transaction Processing

Software, on the other hand, primarily runs mission critical systems in industries like banking, airlines and retail. This includes our offerings such as CICS, TPF and IMS running on z Systems, which were primarily reflected in Other Middleware in our previously reported Software segment. Most of it is on-premise, and annuity in nature, and not a growing opportunity in the software market.

What the heck is Schroeter talking about here? Well in terms of CAMSS, he appears to say that the Cognitive Solutions Segment includes analytics, security and social (the A-S-S in CAMSS) and includes the Watson Artificial Intelligence or Machine Learning platform, which is all very strategic stuff. But you'll also notice he says the segment includes Transaction Processing Software. He explains a bit what that is and where it runs but doesn't really explain why transaction processing belongs in Cognitive Solutions. What is cognitive about it? Why is it reported here? For that matter, what's the point of this financial reporting reorganization, anyway? I have theories but first, let's get to the quarter's good news, the financial results:

In the first quarter, Cognitive Solutions generated revenue of $4 billion,which was up modestly year to year. Solutions Software grew, mitigated by a decline in Transaction Processing Software. Our Solutions Software growth of three percent was a significant sequential improvement from the fourth quarter of last year. Contributing to this growth was strong double digit performance in our Watson cloud-based offerings, and acquisitive content. From an offering perspective, the growth was led by analytics and security. In both of these areas, we have highly differentiated strategies and capabilities.

So to paraphrase Schroeter, Cognitive Solutions sales are up a little and might have been better were it not for a decline in Transaction Processing Software. In short, the new stuff is somewhat better and the old stuff they had to stick in here is a little worse, putting a drag on the segment.

It is my belief that Schroeter is dissembling here. If you aren't familiar with the term dissembling, dictionary.com defines it as "to give a false or misleading appearance to; conceal the truth or real nature of." Schroeter is dissembling the true nature of Transaction Software Processing in the IBM business plan and its real reason for being reported in the new Cognitive segment, which is to make the segment's A-S-S numbers look better than they really are.

Transaction Processing Software mainly refers to the CICS transaction manager (introduced 1966) and IMS database and transaction manager products (introduced 1968). Together they bring in billions in revenue and run a goodly chunk of Western civilization, all on z/OS mainframes. Nearly all their revenue - over 95 percent - is profit. CICS and IMS each have fewer than 175 employees, and did not escape the withering March 2nd IBM layoffs. Both products have existed since the late 1960s and have absolutely nothing to do with anything Cognitive, yet they represent about $1.287 billion of the segment's reported sales and ALL of the segment's profit. Yet these are the very numbers that were noted for having "mitigated" the segment results.

The transaction processing numbers appear to be reported here to make Cognitive results look better and Systems results look worse. This is corporate transformation?

Some may wonder how I can claim that Transaction Processing Software was responsible for ALL the profit in this segment? That's certainly not what Schroeter says. Actually, Schroeter carefully says almost nothing and it is difficult to even reckon from his remarks what is mitigating what.

Jumping to the actual reported numbers for the segment from IBM's 1Q'16 10-Q filing, we find total quarter revenue for Cognitive was $3.979 billion, down 1.7 percent with TPS reporting revenue of $1.287 billion which is notably down seven percent from the same quarter a year earlier. But since we know how TPS works, we also know that about 95 percent of that TPS revenue is profit, which would be $1.222 billion compared to a reported total segment profit of $3.262 billion. Do the math and you'll see that while IBM reports the segment gross margin as being 82 percent, the most likely intra-segment breakdown is a 95 percent gross margin for TPS and a 75.8 percent gross margin for Solutions Software - the A-S-S bits.

So rather than TPS dragging down the Cognitive segment, it is actually pulling it up, which is the only reason I can think of why the TPS numbers are there at all.

Now look at the reported segment pre-tax segment profit of $1.013 billion. This number is the gross profit minus a number of adjustments like the acquisitions mentioned in the management discussion note. For the TPS sub-segment, there are likely to be no such adjustments other than separation costs for the 35 or so TPS heads that were cut March 2nd. This is likely less than $1 million from the $1.222 billion TPS profit. So the pre-tax TPS profit of $1.221 billion is greater than the entire segment's pre-tax profit of $1.013 billion. In other words, on an operational basis, all the Cognitive segment profit comes from non-Cognitive activities.

The IBM financial story line for the segment says business is good except for that pesky TPS sub-segment which is dragging things down a bit. But the truth of the situation is that TPS is propping up the segment, the rest of which (the A-S-S parts) are actually losing money ($1.013B-$1.221B=-108M).

Finally, based on IBM's 1Q'16 report, its Analytics (A) is bigger than the rest of CAMSS (the CMSS part) combined. If the growth and earnings of this unit are not as good as we have been led to believe, then it is possible IBM's strategic plan is a long way from contributing to the bottom line.

How honest is this? How transparent? And this is just one segment. What other smoking guns are hidden in these IBM financials?

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.