G-S Research Report on IBM this morning
IBM will be reporting its September quarter results on October 16. The presentation is scheduled to begin at 4:30 PM EST and will be available on IBM's website at www.ibm.com/investor. We expect no preannouncement and we are generally satisfied that IBM will continue to fare better than most technology companies during the extended slowdown in IT spending, gaining market share and mindshare in a very difficult environment, and having the ability to capitalize on a rebound in demand. Although we are revising our estimates for the December quarter and full-year 2003 downward in the face of ongoing weakness in IT demand, we see the quality of IBM's earnings as improving and IBM as continuing to make headway in its key services, hardware, software, and technology markets. Our conclusions are as
follows:
1. WE ARE REVISING OUR ESTIMATES BUT STILL SEE IBM AS BEING IN BETTER SHAPE THAN MOST TO COPE WITH TODAY'S CLIMATE. We are lowering our EPS estimates for 2002 and 2003, primarily based on expecations that IT spending will remain sluggish at least through the first half of 2003. In addition to weak IT spending, we expect a drop-off in IP trasaction-based royalties as well as incremental pension-related expenses to negatively affect earnings by a total of $0.10 in 2002 and $0.40 in 2003. On the other hand, our revised model shows improved profitability in IBM's microelectronics business and slightly more savings from restructuring, boosting EPS by about $0.19 in 2003. In June 2002, we estimated that restructuring would provide a $0.40 benefit to IBM's 2003 EPS, while an adjustment to IBM's 2003 pension plan return assumptions could reduce earnings by $0.14. The changes that we are making now are incremental to our earlier assumptions. One of our key conclusions is that IBM's restructuring efforts should more than offset changes in the expenses associated with IBM's pension plan. Our new EPS estimates for 2002 and 2003 are $3.90 and $4.30 versus our old estimates of $4.00 and $4.50, respectively. In the table below, we have presented a summary of the effects of each of these factors through the end of 2003, arriving at our new quarterly estimates.
2. ALTHOUGH BUSINESS IN THE U.S. SEEMS TO HAVE GOTTEN PROGRESSIVELY WORSE OVER THE COURSE OF THE QUARTER AND EUROPEAN DEMAND IS DETERIORATING, WE EXPECT IBM TO COME VERY CLOSE TO MAKING ITS NUMBERS. In a quarter that was probably more back-end loaded than usual, many data points indicate that end-of-quarter closings were even more difficult than normal, particularly with large projects. While IBM is obviously a participant in what is a tougher environment than even we would have presumed, it is gradually increasing its inroads in each of its primary areas of focus - services, software, hardware, and technology. Once again, the diversity of IBM's revenue streams should allow the company to outperform its peers. While hardware revenue is likely to be down somewhat year/year, we expect services, software and financing revenues to be flat to up.
3. IBM IS WINNING BY INCHES, PARTICULARLY NOTABLE ON THE HARDWARE FRONT. In the September quarter, we expect hardware to be generally in line with our estimates of around $6.6B, down 3.4% year/year but still notably better than others, with IBM seeing market share gains in Unix servers and storage. Meanwhile, Intel-based server growth, on the back of IBM's scalable, 4-way and 8-way offerings, is strong for the second quarter in a row, in our model up 12-13% year/year.
4. IBM'S SERVICES REVENUES SHOULD ALSO BE OKAY ALTHOUGH BOOKINGS WILL PROBABLY BE IN THE $9.5B AREA. Our modeled services revenues of $8.7B should be okay, flat to up from last year, with the potential for some slight upside variance. Bookings, on the other hand, were undoubtedly impacted by today's climate, with our sense being that signings may have ended up at around $9.5B. Given continued efforts by customers to reduce IT related expenses, we expect IBM's more transactional BIS consulting services to continue to decline (our model shows a decline of 9% year/year) but by a decelerating amount. Highlighting the resiliency and stability of IBM's outsourcing business, we are forecasting mid single-digits year/year growth.
5. MORE IMPORTANTLY, IBM - WITH SOME LUCK ON ITS SIDE - IS NOW IN A POSITION TO BENEFIT FROM EDS's AND TSMC's CHALLENGES. We expect services bookings strength in the fourth quarter. While we had begun to think that the strength of 4Q bookings would be negatively impacted by IT spending weakness-related pushouts, we think issues at EDS have worked substantially to IBM's benefit. Specifically, a number of the deals where the two companies were going head-to-head now appear to be moving in IBM's favor. Even the well-telegraphed Procter & Gamble order could be back in possible play for IBM, this time with terms more satisfactory than the ones we think IBM initially walked from. For the December quarter, IBM may already be well along in potential signings than where it has been in the past several quarters, with a sufficient number of already-won deals that are in the final contract negotiation stage to be close to total September quarter bookings. On top of that, we would expect IBM to be able to win more competitively than it normally might have, bringing total December quarter signings back up to around $14B-$15B. In addition, IBM's first-to-market copper fab should continue to run at near full capacity through at least the middle of 2003, improving the profitability of IBM's microelectronics business. Much of IBM's success in copper appears to come from a lack of competition, particularly at the high-end, where TSMC and others apparently continue to struggle, rather than a pick-up in demand.
6. THE QUALITY OF IBM'S EARNINGS SHOULD NOTICEABLY IMPROVE ONCE THE PENSION ISSUE IS TAKEN OFF THE TABLE. Over the past several years, there have been multiple non-operational items that have affected IBM's bottom line and which have caused many investors to raise flags about the quality of IBM's earnings. Among these items - the reduction of the tax rate from the mid- to upper-30%s to a current 30% level; share buybacks in which IBM typically has spent $5-6B per year; transactional IP agreements such as the one with JDS Uniphase which added about $0.11 to IBM's bottom line in 4Q01; IBM's overfunded pension which was exceeding expected returns and throwing off incremental earnings as a result, including $0.05/share in 2001; real estate sales; and equity investments. Over the past couple of years, IBM has been gradually weaning itself from these contributors and working its way towards a higher quality of earnings. The last of the big items involves pension expense. In an effort to be more conservative and more in- line with other U.S.-based companies, we expect IBM to lower the assumed rate of return on its pension assets to 8.5% in 2003 from 9.5% in 2002 and 10% in 2001. The reduced return expectation will remove about an additional $0.14 or so in earnings in 2003 (we had already removed $0.14 previously when we moved to an expected rate of return on pension assets for 2003 of
9%) and reverse a trend in the late 90s and early 00s when pension had an incremental benefit to IBM's earnings. Although IBM will probably have to reduce its share repurchases temporarily by about $1B to provide the cashflow to fund its now-underfunded pension, FCF is more than adequate to keep its programs virtually intact.
7. IBM IS NEAR THE LOW END OF ITS HISTORICAL VALUATION RANGE. Weak IT spending and a lack of near-term catalysts continue to weigh on all tech stocks. However, despite continuing to outperform its peers and consistently generating strong earnings and cash flow, IBM is trading at roughly 0.8X the S&P500 - at the low end of its historical range of 0.8X- 1.1X. Using our new revenue and earnings expectations, a discounted cash flow analysis yields a 12-month price target of $75, representing 30% appreciation, versus our prior price target of $90. The prevailing risk continues to be the current IT spending environment and the associated uncertainty of an eventual recovery. |
