Outsourcing can be dangerous to a company’s financial health.  That was the apparent conclusion of J. P. Morgan’s COO Jamie Dimon and CIO Austin Adams when they announced yesterday that they have decided to cancel their $5 billion outsourcing contract with IBM in favor of bringing their data centers, help desks, and networks back in house.  According to today’s Wall Street Journal, “Jamie Dimon, the bank’s chief operating officer and designated heir for the corner office, has made it clear that he thinks technology infrastructure can convey a competitive advantage.”  J.P Morgan’s outsourcing contract, first announced in 2002, was one of the largest outsourcing contracts in the industry.

I believe that this is a very significant event that may well spell the beginning of the end of the outsourcing fad that started in the early 1990’s.  My assessment is partially based on the size the deal that was canceled (undoubtedly with significant cancellation penalties), partially on the respect that Jamie Dimon receives from the financial community as being a thought leader, and partially because it makes eminent sense. 

Why does it make eminent sense?  Because a modern bank is at its core is an information processing machine consisting of computers, networks, software, and storage, and employing humans only for those functions that we have not yet figured out how to computerize.  Hence, if you have the best technology infrastructure, you have a competitive advantage. 

You never outsource your competitive advantage if you want to sustain it.  Wal-Mart came to this conclusion in retailing, where it was a lot less obvious than in banking, and has used its in-house proprietary software and computer expertise to run over almost all of its competition.  I am sure that is what Jamie Dimon and Austin Adams have in mind for J. P. Morgan.  The competitors had better start thinking now about how they will counter this move before it is too late, now that the opening shot has been fired.

J.P. Morgan’s and Wal-Mart’s logic is not just confined to the banking and retailing industries.  Computing can be a major source of competitive advantage for many industries.  One of the most obvious is the high tech industry.  Yet, many high tech companies followed the 1990’s fad and outsourced their data centers and networks. I believe that many of them are now suffering the consequences of that decision. Let me give you an example.

In the early 1980’s, when I was a group supervisor at Bell Labs (then part of AT&T and later to be split off as part of Lucent), I gave a presentation that I recently discovered has been remembered to this day.  The purpose of my presentation was to convince the heads of the Bell Labs computing centers that we had to stay on top of leading-edge information technology for the good of AT&T.  My talk was scheduled for the dreaded after-lunch slot of a full-day meeting, so I knew that I had to do something to get my audience's attention.  What I ended up doing was to start off with two slides that appeared to show Wall Street Journal articles dated exactly 10 years in the future.  The  two articles had the exact same headline: “AT&T Files for Chapter 11”.  The first sentence of both articles was also the same: “Today, AT&T filed for protection from creditors under the Chapter 11 bankruptcy act.”  But the two articles differed from there on. The first article stated: “Howard Anderson of the Yankee Group explained the reason for AT&T's demise: ‘AT&T had decided to buy all of their computing and data networking technology and services from others.  As communications and computing converged, they discovered too late that they had developed no competence of their own on the data side of the business.  Competitors who had developed both communications and computing competence and associated  industry knowledge simply ate their lunch.’"  The second article stated this: “Howard Anderson of the Yankee Group explained the reason for AT&T's demise: ‘AT&T insisted on only using their own computers, data networking products, and software.  They therefore became completely out-of-touch with what was happening in the rest of the industry, and the competitors ate their lunch.'"  My presentation had the intended shock value on the audience, who approved my proposal to move ahead with a plan to both procure leading edge technology from market leaders and develop in-house leading edge technology in strategic areas.  The result, documented in more detail in my book, was the development of an in-house live test bed of many of the key technologies that shook up the industry in the 1990’s.

After my departure from Bell Labs in the early 1990’s, AT&T decided to outsource all of its data centers to IBM.  When AT&T spun off Bell Labs as part of Lucent in 1995, Lucent maintained the outsourcing contract, which I believe is still in effect today.  During the 1990’s Lucent became increasingly out-of-touch with computing and data networking driven technologies such as IP, resulting in a near-death experience for Lucent and a substantial reduction in Lucent’s revenue, profits, stock price, employment base, and bond ratings, from which it has barely recovered.  At the same time Cisco, which barely existed at the beginning of the 1990’s, ate Lucent’s lunch and grew to be the largest and financially strongest competitor in communications.

For large companies, outsourcing almost never made any sense other than as an accounting gimmick.  The outsourcing contracts were cleverly structured so that even if the overall cost of computing increased as a result of the contract, CFOs could record numbers that appeared much better to the investor, at least in the initial years of the contract.  Today, with everyone taking a much more conservative approach to accounting, this advantage is not what it used to be.