With HP's third quarter financials due for release in just a few weeks time, it is an opportune moment for a detailed review of the impact of the merger on the respective business sectors and on the company as a whole.
When the merger was announced Compaq was starting to look like a battered wreck. It was a PC company that never stopped being a PC company even after buying Tandem and Digital Equipment. Compaq's PCs had made only a tiny profit in one or two quarters for the previous 3 years and the company was being held together by the parts of the business that they rarely mentioned, the high-end systems that were acquired from the aforementioned.
In late 2000 and early 2001 Compaq's sales of high-end systems hit a brick wall as the Internet boom collapsed and they were caught bare. The division working on the Alpha processor was handed over to Intel as a method of saving money and no doubt Intel would have been happy to get them in exchange for guaranteed sales, albeit at discount prices, of future products.
Rumours from ex-employees claim that Compaq was running out of money and would have lasted less than 12 months had the merger not occurred. Whether this is true is a moot point now but certainly they were not in great financial shape. Compaq's annual statement for 2001 shows that they had borrowed a further $1 billion over that year but in the same time they had made an operating loss of $785 million. Probably more serious were short-term borrowings which stood at $1.7 billion and which would somehow have to be repaid or refinanced in the next few years despite the company making very little profit.
HP's purchase price of Compaq was approximately $24.2 billion and that amount included $14.45 billion for "Goodwill", $3.5 billion for "amortizable intangible assets" and $1.4 billion for "intangible assets with indefinite life". These intangible assets account for about $19.5 billion or 80% of the purchase price and they assign a value of only about $4 billion for the tangible items.
They are also in sharp contrast to Compaq's valuation of the items that they listed as assets in their annual financial statement for 2001 in which they claimed to have total assets of $23,689 million, including one item "Other assets" worth $7,212 million.
Did HP pay too much for Compaq? Well it is difficult to see that the $14.45 billion for goodwill was justified. It must be seriously doubted that Compaq actually had that degree of credibility in the market, especially when it was always regarded as a PC manufacturer, one that couldn't make a profit from its PCs.
Let's take a look at the various business sectors to assess what impact this merger has had on them.
Printing and Imaging
This was the flagship sector of the old HP, the sector that made money while those around it stumbled.
Walter Hewlett recognised that Compaq would add almost nothing to this sector and he believed that a merger would dilute the interest in Imaging and Printing.
Current quarterly revenue in this sector is not dissimilar to revenues back at the first quarter of 2001 and while there was a slump across the middle of that year, the recovery has been happening since the final quarter of 2001 with no sudden increase that one can attribute to the merger.
The merger really appears to have been neither an advantage nor disadvantage to this sector. This is important because it belies the pre-merger claim that merging the two PC divisions would ultimately drive sales of printers and digital cameras or at least be an important part of a "solution bundle" for sales of these products.
From 1999 onwards Compaq had a reputation for selling a lot of PCs but rarely making a profit from them. HP had a reputation for selling fewer PCs but likewise making miniscule, if any, profit.
Walter H was concerned that the economics of dealing in PCs was unattractive because they were a low margin product line and that they presented a greater risk to the business. The pro-merger group made lots of noises about economies of scale as if they somehow counteracted these arguments.
HP's financial statement for Q4 of 2002 kindly informs us that in total the two companies made a loss on PCs in every quarter from Q1 of 2001 and that the total loss was $1.507 billion. The last two quarters in this period were as a combined company and the loss there was more than $250 million.
The situation 2003 is a slight change of fortune with a small profit being generated on high revenues. The Q2 figures show that PCs produced $5124 million in revenue but a profit of a mere $21 million. The PC revenue was not far short of that from the Printing and Imaging division, which by contrast managed to return a profit of $918 million in the same period, which was better than 40 times the percentage return on PCs.
It has been claimed that making a profit from PCs was all about economies of scale, something that apparently Compaq was unable to achieve despite the number of units they sold. The truth is rather simpler; it was about reducing the number of customers for the makers of PC components and thus forcing them to reduce their prices.
Reducing the price that HP paid its suppliers and reducing the number of employees dealing with PCs have enabled have enabled HP's PC sector to generate wafer-thin profits.
The profits on PCs are still small because despite the revenue, the costs of doing business are very high. HP continues the struggle in this sector against Dell and against white box manufacturers with lower overheads. Sure with high revenue and high turnover it only takes a small increase in margin to make good profit but the signs are that HP are going to be struggling to find even that small increase.
On balance, the acquisition of Compaq's PCs has only trimmed losses in this sector. Trimming losses is all very well but being in business is all about making profits and in this regard it has been a failure. With tiny profits and low return on investment, the Personal Systems Group continues to be what it was in the individual companies, not much more than a job creation scheme for its employees.
The merger gave HP a problem with too many flavours of Unix. Compaq's Tru64 Unix was good technically but not favoured by customers while HP-UX was somewhat the reverse. The solution they came up with was to merge them and reduce the costs associated with their development and support.
Much work has gone into reaching common ground between Tru64 and HP-UX but that work is not complete the big question is will HP's customers accept the result given that changes have been forced on all of them.
Unix did very well from the Internet boom because it powered the majority of servers but the bursting of the bubble meant that the demand for Unix-based systems was drastically reduced.
More recently many customers are finding that Linux presents an attractive alternative and HP have been forced into the position where they offer Linux as an alternative to their own Unix. HP is now actively promoting Linux as a string of announcements in just the last week will testify.
The origins of clustering are with OpenVMS and it was the transfer of these ideas that gave Tru64 one of its major technical features. Clustering of a kind is now available in Linux and HP is actively promoting this to the detriment of their potential Unix market, or for that matter their OpenVMS market.
HP now proudly claims that it is a leader in the Linux market but they forget it is a market with small margins.
Walter Hewlett was concerned about HP getting more involved with low margin, commodity products but I suspect this would have happened with or without Compaq, it was just that Compaq's people made it happen a little faster.
It is doubtful that the acquisition of Compaq's Unix systems brought any major benefit to HP. Sure HP-UX was not doing well after the Internet collapse but neither were HP's competitors in this area and this was never sufficient justification for obtaining Tru64 Unix. As things stand, the resources being pumped into the merging of Tru64 and HP-UX could be largely wasted as Linux gains more ground and HP continue to promote its use.
OpenVMS and NSK
For these products little seems to have changed from the days of Compaq's ownership, a fact which has disappointed supporters of both products and who believed that the positive words from HP's executives were going to be followed by positive actions.
If rumours are to be believed, there was no sales boom in NSK and OpenVMS during the Internet bubble and subsequently no bust to drive down the sales numbers. Revenue from these products has been relatively stable, quite unlike Unix systems that have suffered from a major collapse in sales.
Many would argue that the beneficial aspects that have arisen from acquiring these products as a result of the merger appear to be minimal because their potential has largely been ignored. They see these products as offer features unmatched in Unix or Windows but HP's attitude that they are legacy products that are only suitable for niche markets and not worth the effort of marketing them to the wider IT market. Of course the continued management of these products by those executives responsible for them at Compaq has meant that their relative invisibility to potential customers has continued.
There are good reasons to suppose that OpenVMS and NSK were acquired with some reluctance as products themselves and only a little more enthusiastically regards the support and service contracts that came with them. These contracts are slowly diminishing because customers see that HP's reluctance to market them has followed Compaq's reluctance. There is little enough knowledge of these systems at a CIO level and even less confidence in their long term future at HP.
HP's services sector appears to have gained from the merger with Compaq. It is not so easy to be definite because outsourcing is a very fluid business which is currently experiencing an increase in demand, particularly to those outsourcing companies who have access to low-cost IT workers.
The services business has always been low cost because there is no R&D and no manufacturing to eat into the revenue. All it takes is a body of IT professionals, usually a combination of employees, contractors and extra staff acquired from the customer companies. In short it only relies on the credibility that comes with having a large number of staff and the ability to maximise profits by optimising the costs of those staff members.
It was Compaq's service sector that HP seemed to regard as the glittering prize in the merger. The acquisition of the services division picked up a number of existing contracts with their staff and cash flows. Most of these service contracts would have been for Compaq's own products, namely Tru64, OpenVMS and NSK, and with a fair sprinkling of support of Microsoft products that together made it a generally useful collection.
By virtue of reduction in the number of competitors and an increase in credibility that came from an increase in size HP has picked up additional contracts as the demand for outsourcing has increased. Good access to low-cost staff has enabled HP to reduce overheads but as we have seen recently this lower-cost has sometimes meant lower quality and very unhappy customers.
So how well is this sector doing? In 2000 and 2001 Compaq's services arm was returning about $250 million per quarter and HP's about $120 million. In Q2 of 2003 HP's Services sector returned income of $301 million, which is about $70 million less than one would have expected.
HP has recently announced a number of major outsourcing contracts such as the 10-year, $3 billion contract with Proctor and Gamble but this means $300 million in annual revenue when HP's services sector currently has annual revenue of about $12 billion.
From the size of Compaq's contribution to this space it appears that this Services sector was probably presented the only serious compelling argument for HP to acquire Compaq. Moreover it is probably the only area in which HP would have had competition if the stock price of Compaq had fallen further to the point where there was a "fire-sale" of Compaq assets.
No discussion of the merger would be complete without mention of staff retrenchments because this reduction of overheads is at least as important as the sales of products and services.
It is characteristic of most mergers that it is seen as a good opportunity to retrench workers. This usually goes beyond the reasonable premise of removing duplicated job roles to become a clean-out of other positions in a manner which might meet strident opposition had it been done at another time. The mantras of "improving efficiency" and "reducing overheads" are used and they accept no dissent.
HP has certainly done this and admit they have retrenched more people than was originally planned, an admission that suggests that HP and Compaq had excessive staffing levels prior to the merger. If this is the case it raises interesting questions about the mismanagement got them into that situation in the first place.
It is axiomatic that had Compaq collapsed even further and HP not acquired some of the above product lines not been acquired, then the staff that accompanies those products would not have been transferred to HP and yet HP would have stood to gain as they filled the vacuum left by Compaq's departure from the market place.
Has the merger been successful?
The pre-merger talk was all about synergies and cost savings and how Compaq's products would fill deficiencies in HP's product line.
Those cost saving have materialised through the retrenchment of staff and the ability to pressure component manufacturers into reducing their prices, but the positive synergies of acquiring Compaq assets remains very uncertain.
The second quarter of this year saw HP's earnings from operations at $1210 million. Compaq's financial quarters were out of synch with HP's but using the closest calendar dates the aggregates for the same period in 2000, 2001 and 2002 were $1468 million, $885 million and $488 million.
HP's earnings are up overall but whether that is directly due to the merger or the mild recovery of the market is more difficult to determine.
Growth in Printing and Imaging has continued and it returns good income (up $150 million this year over last and a massive $550 million over 2001) and it accounts for much of the increase in total earnings.
The income from PCs and high-end systems has improved and is now near the break-even point. PCs have made significant losses for the separate and merged companies over the last 4 years and the merit of selling them in the current manner continues to be dubious.
If the steps implemented since the merger have not resulted in decent levels of income from PCs then it is even harder to justify the continuance of selling these products. Boasting rights that come from having the highest revenue are all very well in theory but unless are profits are accompanying that revenue it is all rather pointless.
The profitability of HP's services sector has also increased since the merger, although by not to the extent that one might have expected. The service sectors of the two separate companies returned a total of $392 million for Q2 of 2002 and $428 million in 2001 but in Q2 of 2003 the combined services sector returned only $301 million, which was close to a 25% fall on the total for the matching quarter of the previous year.
Against these overall improvements in income there have been significant costs of restructuring with the retrenchment of staff and the closure of various buildings and facilities, and of course here is the ongoing liabilities of retention bonuses for executives and other key people.
The merger may be producing increased income at the moment but whether that will continue against the pressure from the increasing use of Linux and the pressure that is driving down the margins in the services sector is anyone's guess. But of course these issues were visible back at the time that the merger was announced and they should be no surprise to HP.
All up, regardless of whether the purchase was viewed a defensive action against competitors or as a means to increase profitability I doubt very much that HP got good value for the $24 billion that they paid for Compaq.
I also suspect that HP would have done better to have allowed Compaq to drop even further in value and perhaps implode, and then either pay a lower price for the whole or only acquire the only pieces that it wanted. Either way HP could have outlaid a whole lot less and still got what they wanted.
There were few, if any, competitors who might have attempted to buy Compaq and they would have only been interested in specific components.
I also don't believe there was much risk that Compaq's stockprice was going to rise significantly over the subsequent 6 months or even 12 because they were fixated on low-margin products which suffered from strong competition from companies with fewer overheads.
The marketing exercise that sold people on the idea of the merger extolled the many benefits that would result. Personally I cannot see that those benefits have materialised to any significant extent. Moreover when the costs of the merger are taken into account and the payback period is still many, many years away one must question its wisdom. µ
Getting botter all the time
It's the best of the rest from Google's week
Just like we promised ourselves we wouldn't do again