A
recent Deloitte
survey of 25 of the world’s largest business organizations showed that 64
percent of them had brought outsourcing deals back in house. Another, from Dun
and Bradstreet, reported that 20 percent of outsourced relationships fail by
the second year, with 50 percent failing by the fifth year. In addition, PA Consulting
results indicate that 15 percent of companies were thinking of bringing
outsourced deals back in house.

What
is happening to outsourcing? Is the practice of moving business functions out
of house falling apart at the seams? If you have been watching the headlines in
recent months you might think so.

The financial services
sector looks particularly vulnerable. Several megadeals have, apparently,
fallen apart. For example, JP
Morgan Chase, following its merger with Bank One, withdrew from an outsourcing
arrangement with IBM to manage its entire IT infrastructure—a $4.8 billion
deal over 10 years. The bank’s new leadership expressed a preference for
in-sourcing.

Tips in your inbox

Stay up to date with the latest IT news and information affecting the world of finance with TechRepublic’s free Financial Services IT newsletter, delivered each Wednesday.

Automatically sign up today!

In
addition, UBS has announced that it will bring its outsourced infrastructure
and market data deal with Perot back in house when it expires in June 2006. The
reason given is that the bank is changing its business strategy, which includes
centralizing IT across the group. Likewise, Norwich Union, the world’s sixth-largest insurance
group and the biggest in the United Kingdom, has cancelled a network
outsourcing deal with IBM worth several hundred million dollars because of
a change of business strategy after a merger.

But
it would be wrong, I think, to see these headlines as symptomatic of a wider outsourcing
malaise. First, the reasons for the return to in-sourcing in these cases all
share something in common: they are a result of being able to achieve the
economies of scale in-house that were originally sought in the outsourcing
deals. In the cases of JP Morgan Chase and Norwich Union, the in-house
infrastructure volume has been achieved as a result of mergers. In the UBS case,
the economies of scale have been reached because of an internal global
consolidation program.

Second,
although these in-sourcing decisions are high-profile, they represent a
fraction of the number of outsourcing deals that exist in the financial
services IT sector. Even if you count only the very largest BPO (business
process outsourcing) arrangements, the truth of the matter is that most of them
work, and work rather well. There may, of course, be periods of renegotiation
and adjustment. But then, periodic break clauses, which provide a deadline for
such assessments, are regarded as best practice.

Moreover,
even if mergers and the like mean that economies of scale can be located
internally, there are reasons to resist the in-sourcing route.

It
is risky to switch course: Having outsourced skills and systems means that you
have to re-source them—and if that
means bringing people back in-house, there can be serious questions of morale
to address. It is also time consuming; if a BPO deal took 18 months or more to
put together, it is likely to take almost as long to undo.

The
shift from outsourcing to in-sourcing may also prove to be premature. When
in-sourcing is pursued as a means of driving down costs, it is worth bearing in
mind that IT costs are also decreasing all the time too, and it is likely that
infrastructure providers are best placed to realize these savings fastest.

So
the next time you see an in-sourcing headline, look at the small print beneath
it. More outsourcing, not more in-sourcing, is still the underlying trend.