Wanadoo just reported its H1 2003 results and the FY 2003 Group EBITDA target looks well in hand thanks to the outstanding performance of the directories division. The performance of the Internet segment has been less satisfactory for two reasons: Wanadoo France is facing stiff competition from Free on the 512k DSL segment; and Freeserve in the UK and Eresmas in Spain have seen very slow subscriber and revenue growth due to barebones customer acquisition activity. Wanadoo will be ramping up DSL customer acquisition activity from September onwards to achieve Internet segment targets and may reduce prices in the UK.

Wanadoo reached an important milestone in 2002, reporting its first (very small) positive EBITDA margin on its French Internet business, thanks to broadband-related revenue increases and lower narrowband and broadband access costs. In contrast, losses widened at Wanadoo's Internet properties outside France, in particular Freeserve in the UK and Eresmas is Spain, but these were more than fully offset by profits on the Directories segment. This note looks ahead to 2003, when Wanadoo expects to reach positive EBITDA on the Internet segment as a whole, thanks to continued improvement in France and tightly contained losses at Freeserve and Eresmas.

Wanadoo's results for the first half of 2002, detailed in the attached note, show that the company is well on track to make its target of positive EBITDA as the loss margin has been cut by half on the Internet side of the business. The targeted revenue increase of 30% also looks plausible as Internet access revenues have done well in France due to migration of the subscriber base to higher priced broadband packages. Wanadoo hopes to have 1 million broadband subscribers by the end of the year, and is counting on the rollout of a new lower speed (128k) and lower-priced broadband package in mid-October. The French Competition Commission has also permitted the company to again market its broadband packs in FT's network of shops, cutting customer acquisition costs. Margins will improve in mid-October due to wholesale broadband price declines mandated by the regulator ART.

We think that the business is worth about €6bn, rather less than the €7-9bn that the investment banks are projecting. The difference arises because we think that they over-estimate the value of Universal’s music publishing business and expect a faster upturn in recorded music sales. But Universal is clearly strongest of the major music companies and we do expect the company’s margins to recover from the low levels seen this year.

This note contains our latest update on Wanadoo, France's leading ISP and broadband service provider, following on from the report we issued in April. Wanadoo's Q1 2002 results are on target with the company's objectives for the year, despite sharp declines in portal and e-commerce revenues. The reason is Freeserve: a better deal from its network provider has raised ARPU to €5.7/month from €3.7/month in Q4 2001, and its PAYG customer base has expanded under continued marketing efforts.

Wanadoo is a business combining extensive interests in European ISPs with a strongly cash-generative directory business. Wanadoo's position as the leading French ISP is secure. Its position as an ISP in other markets is much less happy; in particular, Freeserve in the UK is not performing well. In this report, we address the underlying reasons why the French ISP business is healthy while the low ARPUs and poor or negative access margins in other countries are draining the company's profitability. Section A of the report provides detailed projections of 2002 for Wanadoo ISP operations. We try to show why the unmetered access model for narrowband ISPs is dangerous.

This note inquires into the difficult question of what really drives the capital expenditure of mobile operators. We try to show that since much capital investment is actually replacement of existing assets, the importance of the declining growth rate in call minutes in reducing capex is overstated. Our - very rough - estimate is that a mature European 2G operator will probably have to spend about 15% of sales on capital expenditure for years to come. This is in marked contrast to the more optimistic operators, who have publicly offered targets of below 10%. Similarly, we see little relief from 3G. While it is undoubtedly true that 3G provides more bits per buck, the costs of running a 3G network alongside a 2G infrastructure more than outweigh this advantage. Observers should also note that the capital efficiency benefits of 3G are largely illusory, since the savings in the network are wiped out by the higher handset costs.