The ins and outs of up and down

Does the push to consolidate IT services across the federal government mean administrative civil war? Maybe; maybe not.

This much at least is certain: The bureaucracy today works inside vertical structures that have been refined over 30 centuries to resist interference. Meanwhile, the economies and utility of common administrative infrastructure make for an irresistible horizontal force.

That horizontal force earned a vivid description in October from Jim Alexander, acting Chief Information Officer with Treasury Board Secretariat, in a presentation at the GTEC technology exposition. Alexander observed flatly that “the issues of today and tomorrow are horizontal, and, to respond to these issues, departments and agencies will need to act horizontally – in other words, as a single enterprise.”

In that view, the key to reduced head counts, economies of scale, improved service and better transparency lies in a smooth and efficient transition. But the barriers are formidable and there is no guarantee of success.

The case for common services

The biggest challenge is information. The federal government simply does not have the necessary information about its own activities to run them properly.

Speaking in March to a seminar on government, Treasury Board President Reg Alcock put it this way: “I keep asking how do you manage something you can’t understand, you can’t visualize, you can’t get a sense of?”

Along similar lines, Alcock told delegates to the Lac Carling Congress in May that politicians and bureaucrats lack the tools to manage the business of government. “It would be nice to know how many people work here. It would be nice to have one HR system in the federal government. That’s not rocket science, right?”

Building a solid business case for transformation is difficult if not impossible. Last January, a report by the Information Technology Services Review said it had “… struggled with the intractable problem of determining how much is expended on IT services across the federal government, what it is being spent on, and what the actual outcomes of that spending might be.”

The report conceded that taxpayers might find it “somewhat incredible” that $5 billion a year cannot be tracked, but “departments have each developed their own ways of organizing IT services and their own nomenclature for describing them. The resulting variance in what is tracked and how it is measured makes it virtually impossible to roll up and compare IT cost and performance data across government.”

Internal obstruction

The greatest obstacle to change may be the public service itself. Most senior managers have succeeded within free-standing structures, but now are tasked to be the agents of their own horizontal transformation, changing business processes to conform to common standards, surrendering head counts and resources and becoming more fully accountable. Speaking with at Lac Carling in May, Alcock said, “Money is not the problem. The problem is internal management consensus. These guys are fighting me every step of the way, even some of the smart ones, not because they are personally resistant to it, but it just represents such a threat and loss of control.”

A widely circulated presentation by former Deputy Minister Harry Swain in November to Assistant Deputy Ministers in the federal government illustrates the attitude some senior managers may take to shared services. In a year or two, Swain said, a new deputy would “… discover that your critical head office functions have been compulsorily outsourced, without competition, to a common service agency with no track record but a monopoly on your business nevertheless.” Are data centres, desktop computers, IT security and other services really “critical head office functions?” If they are, is it worth additional costs of 15 to 20 per cent to keep them there?

Unwillingness to change may spur some senior managers to provide facts and figures that have previously eluded them, but they may not necessarily justify shared services. As the IT Services report noted, “Some large departments/agencies that already enjoy substantial economies of scale may resist participating in common service delivery, thereby significantly undermining the savings and other benefits that can be realized by the government as a whole.”

Follow the money

A number of possible structures for common service organizations are on the table, and money is the key factor – speedy investment will bring faster payoffs. The Crown corporation model, and access to private capital, could bring government-wide implementation of first-wave common services in three to five years. Mandatory implementation and fee-for-service funding would probably take six to eight years; it would take longer for departments to reduce costs appreciably.

As Ottawa IT consultant John Davis told a recent CIO Government Review event, “One of the challenges that we see from the centre or with the centre is that there is an enterprise agenda, but there is insufficient enterprise funding and commitment, and a lot of the benefits that people put on the table for shared infrastructure and for enterprise solutions are benefits that are really benefits to the enterprise and rarely easily transformed into business benefits for line departments.” However, line departments have the money, “So we have this conundrum that there is a big benefit to the government of Canada if we do things as an enterprise, but all the money and decision-making is really driven at a program level, and not even at a departmental level.”

Free will

A recent Accenture study of shared services initiatives around the world identified a “lack of committed leadership support” as one of the top challenges to success. When it listed high-risk issues, the IT Services report put leadership at the top and linked it with past failures: “A poor GoC [government of Canada] track record for achieving transformations of this magnitude – requires strong political will and leadership from the top.”

The services revolution could also bring political pushback from every federal Member of Parliament, because the price of lower cost and better service to Canadians will be a measurable impact on local employment levels. Cutting 41,000 jobs means an average loss of 133 jobs in each of 308 federal ridings. Urban centres would probably be hardest hit; ridings in government hubs like Toronto, Ottawa, Montreal and Halifax would be hit even harder.

The government marketplace

The government of Canada has more than 400 “lines of business,” which makes it a complex organization but does not make it a business. Horizontality is a way to achieve private sector efficiency, but at least one critical success factor is missing. After surveying several IT projects, Davis concluded that they succeed when there are sufficient resources – which governments have – and a critical business need – which may be missing.

Alcock told CGR that “large private sector organizations went through this in the late 80s, and they were driven by crisis. There is a sense that the private sector does this rationally. It doesn’t do it rationally; it does it because it is compelled to do it.” Profit motive and fear of bankruptcy do not exist within government, so as presently envisioned, common services would be mandatory, even though the IT Services report noted that “monopoly service providers tend to become less responsive and more inwardly focussed over time whereas competition fosters an ethic of client service.”

Shared services may never compete directly with the private sector, but at least one product offering has adopted many private sector approaches to spread throughout the federal government. RDIMS – the Record, Documents Information Management System – grew out of the Shared Systems Initiative in the late 90s as part of a strategy for improved program delivery and cost savings.

The idea was to move departments to a smaller number of record management systems using a common, cooperative approach to system and services acquisition. RDIMS moves the burden of record-keeping from information managements specialists like archivists and librarians to everyone who manages information. For a government struggling with compliance, transparency and accountability, an efficient enterprise-wide records management system is imperative. As one senior federal executive said, “It really is core because it is the only way we are going to get a government-wide information management system.”

In early 2004, Treasury Board gave the green light to buy more than 200,000 RDIMS licences, with an initial target of 100,000 licences as “break-even/value point” by April 2007. Relaunched like a commercial product in September 2004 RDIMS is a close partnership with CGI Inc. and reports to a board of directors. The program has its own marketing and client services managers and, like any major private sector software offering, user groups to make sure that successive iterations really do fix problems and incorporate wanted features.

As with any product, RDIMS needs to be sold, so the program has what is called an “aggressive client engagement/marketing strategy;” the sales force can wheel and deal with incentives like cost sharing on training and site readiness requirements.

As a central, standardized, searchable information bank, RDIMS itself could become an important tool in transforming government. It provides practical horizontality to departments that must work together to deliver their programs so it has a built-in “network effect” – like fax machines or e-mail, the more departments that use the system, the more benefit there is for everyone.

Like Secure Channel, RDIMS represents a major upfront investment, one that will only pay off when a critical mass of departments and agencies begin using it. The more successful RDIMS becomes, the more it will contribute to the image of the new ITSB as the pre-eminent enterprise shared products and services provider to the Government of Canada. There is no question that the government IT community and private sector vendors are watching RDIMS to see if it can provide value for money – but, perhaps more importantly, to see if the model of a well-funded sustainable business can work for other products and services. In other words, it is a flagship product for a new brand.

The RDIMS project has many of the trappings of a private sector offering, even down to the presence of a central agency ‘”investor” looking for a return, but there is one major difference. If it cannot overcome sales resistance by enhancing the product, slashing prices and supporting a solid community of users, RDIMS can be made mandatory

Inside from the outside

The IT services review listed some excellent reasons for large-scale outsourcing: Less need for investment capital; it would be a fait accompli in terms of forcing change; it would be a straightforward way to acquire new skill sets; vendor firms would get the project management headaches. However, outsourcing is simply not on the table because, among other things, it would require “workforce adjustment” and vendors would get “some or all of the available savings.”

Regardless of the respective merits of these arguments, the decisive factor in the decision might be the Minister. Alcock made it clear to CGR that, if a solution works in the private sector, it should work in the public realm as well. In the past, he said, privatization meant, “We cut off chunks and threw them out. . .

‘Innovative, free from the constraints of government, they need to be able to ride the innovation trail and service local communities.’ Then I thought, if those things are good for them, why aren’t they good for government?”

Shape of things to come

IT common services could arrive in a variety of organizational shapes and sizes, including Crown corporation, Special Operating Agency or “start-up.” The IT Services Review appeared to favour a new organization over a reconfigured ITSB, because it will “… likely prove more acceptable to wary departments, increases the likelihood that a new management team that is unburdened with previous history can create the necessary highly responsive, innovative, client-centred working culture, and provides a level playing field for the ‘best and brightest’ in the GoC IT community to compete for the new positions, both technical and managerial.”

Whatever shape new common service organizations assume, they face a hard battle to create a single enterprise within a culture of organizational isolation.

Richard Bray ( is an Ottawa-based freelance journalist specializing in high technology and security issues.