You will be redirected to the page you want to view in  seconds.

To make IT progress, challenge the conventional wisdom

Aug. 8, 2014 - 01:27PM   |  
  |   Comments
Robert Gates, Principal at RiceHadleyGates and former defense secretary, speaks during the Federal Innovation Summit 2014.
Robert Gates, Principal at RiceHadleyGates and former defense secretary, speaks during the Federal Innovation Summit 2014. (Rob Curtis/Staff / Staff)


Robert Gates stepped down as secretary of defense in July 2011 after almost five years at the helm. In that time, he overhauled dozens of DoD programs and organizations and earned a reputation as a bare-knuckled reformer.

Speaking to an audience of mostly federal IT managers at the Federal Innovation Summit in Washington, D.C., in July 2014, Gates ó now a principal at consulting firm RiceHadleyGates ó offered insights into his internal battles to bring innovation and efficiencies to the DoDDefense Department. Following are edited excerpts:

As CIA director in the early 1990s, one of my chief management tasks was dragging the agency and the other intelligence organizations into the post-Cold War era, which happened to coincide with the emergence of the information age. At that time, each of CIAís four major directorates had its own computer system, none of which could talk to the others. Presumably, things have improved since. Despite our, at times, maddening technical backwardness, the agency did play an important role in the development of the American information technology industry. CIA was the first customer of a guy named Larry Ellison, and the project they worked on was called Oracle. CIA, and NSA for obvious reasons, pioneered the world of data storage and retrieval. But by the mid-1980s, we knew that the private sector, with its speed of innovation, was far outstripping us. And we began at that time relying on commercially available technology and software.


Gates: Agencies must break bad habits to modernize IT systems

Cost savings from IT reform are ... nil?

3 keys to successful desktop virtualization

DISA applies Big Data analytics to better secure DoD networks

Take a measured approach to private clouds

Later, when I was president of Texas A&M, it seemed like every college and research lab had its own computer and email system, each complex and expensive. So, Iím well aware of the challenges that you face as government IT managers ó challenges made all the more difficult given the impact of sequestration and budgetary uncertainty. On that note, I wish I could say that our nationís elected representatives will overcome partisan dysfunction and the physical uncertainty that looms over the federal government, especially the critical agencies and operations funded by discretionary spending. But given the unlikely prospect of that hopeful scenario occurring ó at least in the next few years ó all federal components are going to have to make due with flat or declining budgets and with significantly less resources to invest in the future, including in IT modernization.

At the Pentagon, I tried to make a virtue of necessity by using the prospect of much tighter budgets to undertake reforms on programs acquisitions, as well as on overhead ó reforms that should have been happening regardless of budget levels. The same dynamic is at work in trying to reduce federal IT costs while improving performance and protections.

In the end, I believe that what is not necessarily required is knowledge of what to do, because, in many cases, the technical solutions have been around for some time. As leaders, what is needed is a faster and more streamlined decision cycle, judgment and money, and recognition that doing more of the same is unlikely to achieve better results.

What is needed most of all are leaders who are prepared to challenge conventional thinking, break crockery, stop doing what doesnít work well or at all, and set a new course ó all of which, of course, is the opposite of what goes on in most bureaucratic structures. But radical departures can be done, as we showed in Defense in support of the troops on the front lines in Iraq and Afghanistan, where acquisition decisions and delivery of equipment to troops often was measure in weeks and months, and not the usual years.

Seeing the bleak physical outlook ahead, I spent my last two-and-half years as secretary trying to prepare our defense institutions ó [which were] accustomed to the post-9/11 decadeís worth of no-questions-asked-funding requests ó for the inevitable flattening and eventual decrease in the budget.

This entailed creating as much headroom as possible under the existing topline to protect the military size and technological superiority. The first part entailed canceling or curtailing some three-dozen poorly performing and excess weapons programs, more than $300 billion worth across their projected life span.

The second was to reduce the departmentís overhead expenditures. The goal was to reinvest those savings in force structure and other key combat capabilities.

In 2010, I announced a set of initiative aimed at slimming the departmentís back office, and improving efficiency across the department as a whole. Reforming the Pentagonís approach to IT, which then cost about $37 billion a year, had to play an important part in that efficiencies effort. We found that most of our bases and headquarters had their own separate infrastructure and processes, which drove up costs and created cyber vulnerabilities.

Like much of the federal government, the Pentagon planned to consolidate hundreds of data centers and move to a more secure enterprise system [which was] estimated then to save more than $1 billion a year.

In 2010, we were only beginning to think about the cloud. When it comes to major government IT projects, history is not encouraging. The Air Forceís attempt to role out a new computerized-logistics system [Expeditionary Combat Support System, or ECSS] was an expensive flop. The joint project called DIMHRS [Defense Integrated Military Human Resources System] was supposed to unify personnel systems for all four military services. Several years, and more than $1 billion later, I canceled it.

One of my chief regrets as defense secretary was the Pentagonís inability to come up with a common platform for sharing electronic medical records with the Department of Veterans Affairs, which continues to be the cause of much delay and dysfunction we have recently seen.

I finished my time as defense secretary not satisfied with the progress made in the overall efficiencies campaign.

With respect to IT, the challenges and costs continue even as the Pentagonís CIO rolled out a new IT modernization strategy, and OMB [the Office of Management and Budget] pushes ahead with the Federal Data Center Consolidation Initiative.

Addressing today's IT challenges

The challenge here, as with many areas of government reforms, is that IT initiatives that can generate major savings over time typically require a significant financial investment upfront [which is] all the more difficult in an era of sequestration. Todayís problems also require more collaboration, cooperation and consolidation across organizational boundaries. The technology can become obsolete or the costs increased dramatically. And the sclerotic federal contracting system is not a good match for the fast-evolving information world. At todayís speed of change, government finds itself exactly where the CIA was in the mid-1980s. It needs to turn for expertise outside of government.

Another painful lesson when it comes to major weapons programs and acquisitions: We found that attempts to create wholly new systems and structures from scratch ó the traditional approach to defense procurement ó rarely ended well. The Armyís Future Combat Systems program comes to mind, as do many other examples.

The tough lessons of recent years have forced government leaders to look at modernization, especially with respect to data storage, to focus less on costly and complex consolidations and more on measurable, tangible steps to reduce the costs and improve performance. As in the rest of the government, new thinking and innovative technology are required to achieve the savings that OMB believes are possible in the IT realm.

Government agencies will have to collaborate more instead of approaching technology in a stove-piped and fragmented way. Turf fights, parochialism and the demand of each organization exclusively to own its IT all must give way to new thinking and to new approaches. That will take sustained leadership and management discipline.

(Audience Question): I was wondering if you had any thoughts on how the numerous efficiencies measures that started under your tenure have fared under the changes in leadership that have taken place since you left.

We recognized at the outset that implementation and follow-through was going to be critical, because the changes that we identified, in many cases, extended over a 10-year period.

I would say that there were two pieces to the efficiencies. The first was something that I think was unprecedented in the realm of budgeting and budget cuts. I told the services that their bogey, their goal, was to cut $100 billion all together in overhead costs over that 10-year period. And I told them that, if they identified overhead cuts that reached that goal, that they could then turn around and prioritize additional military capability acquisitions theyíd like to make, and that Iíd give that money back to them. So it was a unique budget-cutting exercise in the Pentagon in the respect that what they identified as cuts in overhead Ė in tail, if you will ó they had the potential to get back as increased tooth.

The second part were the efficiencies in the department itself. That included IT. Those were the ones where we actually cut the top line. The first year we returned about $80 billion to the federal budget over a five-year period. So the Department of Defense, just for the record, by the time I left, had cut about $500 billion from its budget over a period of 10 years: The $330 billion in weapons programs cancellations, or caps, that we put on in 2009, about $180 billion in cuts in the efficiencies exercise in 2010.

Then in the spring of 2011, the president said, ĎThatís a great job, why donít you do it again?í What started out as another $400 billion cut ended up being $485 billion. So thereís a long answer to your question, but then when you add sequestration on top of that, my guess would be that the $100 billion that I gave back to the services for investment in military capabilities has evaporated. Iíd be amazed if the services actually saw any of that money.

What was done in the efficiencies [initiative] in 2010 in the department as a whole was just a down payment on the changes that were going to have to be made to accommodate the additional cut over a 10-year period of $485 billion that was mandated by the president, not to mention the sequestration dollars.

So I think what has happened is that my effort to try and anticipate the budget cuts, to get in front of that train, was incredibly naÔve Ė in thinking that when that train really got rolling, having done the right thing and having shown that the department actually could cut its own overhead, could get more efficient, was going to carry no weight whatsoever. So my view is the savings that were going to be returned to capabilities are gone. And the savings that we identified for reducing the top line have had to be multiplied many times at great cost in terms of readiness and capability.

Can you talk about your experience now having left the federal sector and going into the private sector? What sort of lessons learned have you taken away?

What Iíve seen in private sector fortunately has been largely good experiences of innovation of investment and of continuing to innovate. I think that part of the problem in federal government is that innovation is cyclical, partly because of the budgeting process, partly because of the difficulty of the acquisition process.

And so, you make one change and everybody kind of wipes their brow, and says, ďWhew! That was really hard!Ē And then, everybody sits back until a new generation of capability is required. And so, what I donít see other than in some of the federal research labs like DARPA and perhaps others in other departments, I donít see the continuing research and visualizing of the future, if you will, in terms of IT that is on a continuing and evolutionary basis integrated back into the systems that exist within the federal government. Things sort of tend to get done in blocks instead of a flowing river. I think that thatís one of the shortcomings. Figuring out how to change the acquisition process to avoid that, I think, is a major challenge for managers. It really just requires firm leadership at the top of different agencies and people willing to empower those below them to carry out these kinds of functions.

Iíd make one other observation about IT. Here Iím speaking as somebody who has suffered through more PowerPoint briefings than really is humanly tolerable. IT change is always sold on the basis that itís going to save money. Iíve been at this for the better part of 30 years and I have yet to see a single dollar saved. What happens is another good thing, and that is the new capabilities, the new technologies, give you the ability to do more better. And so, you donít necessarily save money. Iíve very rarely seen real dollars saved through change in IT, but what I have seen is extraordinary leaps forward in capability and in efficiency. So Iíd be very careful in promising a significant savings from IT innovation unless you can really prove and are confident that what youíre doing will fall to the bottom line, Iíd be very cautious about it because you will hurt your credibility with the people who you are trying to persuade to support you.

How do you balance innovation with security? You want to be on the forefront, you want to be moving quickly, but in the federal government we have a lot of systems, which are very sensitive in a lot of different ways: personally sensitive, defense sensitive. How do we make sure that we are not moving too quickly to keep them safe?

Yeah, actually Iím kind of surprised to know that we have anything that is still sensitive. It is a problem. I spent my whole career in that world. If somebody were making the presentation to me about a new technology and a new capability, I would say, letís try it in areas where the security isnít as important. Letís make the experimental platform one where security is a secondary consideration. Letís see if it works, and then see what we can do to enhance the security in it.

More In Federal IT