A comment on “Social Security 2020: Vision and Strategy”

The paper Social Security 2020: Vision and Strategy presents precisely what the title says, a vision and a strategy. The paper does not give a plan. A plan was actually in the works, and a lot of progress has been made, but that project was nixed. There are detailed documents inside the Agency that describe this work, there is even some prototype code, and I hope more people get a chance to see those.

The paper has one fundamental insight and five major themes. The fundamental insight is a shift in strategic thinking. Currently, the Agency looks at what it has today and figures out what to do with it in the next few years. This sounds logical, but when it comes to strategic thinking, it is constraining. My starting point is to consider what Social Security has to do by law. If we had to create an Agency from scratch today, what would it look like? Once we figure that out, then we go back to today’s reality and figure out how to get from where we are to the ideal. Not that this ideal would necessarily be the Agency we would ultimately achieve, but at least having an ideal would set a direction.

The first theme is an operational vision, describing what a modern Social Security should look like from various users’ points of views. It describes the types of environments and services that citizens and Social Security employees want and deserve. It is the operational ideal that will drive our modernization direction. SSA cannot attain this vision by continuing to update and augment to today’s archaic infrastructure as it is currently doing, even if it spends a lot more money,.

The second and third themes relate to technology. The second shows that SSA’s technology costs of the past decade have increased so much that these investments are no longer paying off. It also posits that continuing along current trajectory will increase costs even more. The third highlights that modern commodity IT technology (which drives many very large enterprises today) is so much cheaper than what SSA uses and plans to expand, that it makes sense to at least consider changing. The paper does not present a specific technology plan. Before the project was killed, we were well on our way to estimating costs for modern IT supporting Title II programs.

The fourth theme is an architectural vision. Two major engines will drive the operations of the ideal Agency. One is a Customer Relationship Management (CRM) system that facilitates the day-to-day activities of Social Security employees. The other is a rules engine that automatically guides users as much as possible through the complex, arcane Social Security rules. CRM systems and rules engines have advanced tremendously during the past couple of decades, development and standard usage paradigms have emerged, and a lot of experience can be leveraged. The architecture vision has one dataset driving the entire Agency, so data across all services is consistent. Data is structured to meet the special requirements of Social Security. Data is readily accessible to any user who needs it and has access rights to it. The overall architecture will be service oriented, deployed on commodity hardware and leverage open source software when it makes sense.

The fifth theme dealt with what I considered the hardest problem- the strategy. I am leery of planning a massive IT project and then trying to rigidly stick to this plan. We have all seen too many large IT projects fail, and in a government environment of constraining contracts and procurement regulations, the odds of success are even worse. The strategy calls for building and deploying manageable parts. We identify a relatively small but still significant part of the ideal and build it from the ground up, following our architectural guidelines. We allow both new and old systems to operate simultaneously, and the connectivity between them is designed to be temporary. When this part is operational, we build a next part. This process continues, and at some point, we start retiring parts of the old system. This way we are always moving in phases towards the ideal and gradually getting rid of the old. At the end of every phase, the overall efficiency of the Agency and the quality of its services have improved. We were actually testing two possible starting points, building prototypes, before the project ended.

The strategy calls for budgetary considerations throughout the project. And, critically, the strategy calls for flexibility. While it is important to have some initial roadmap of manageable parts, we should anticipate that as we deploy and learn, it may be advantageous to make changes.

That, in a nutshell, is what the paper Social Security 2020: Vision and Strategy is about. It is not a plan. It provides guidance and backs it up with facts and reasonable arguments. Given what we know about the current IT environment at Social Security, what we can expect if the Agency continues with business as usual, I think that investing a little bit on an alternative that has excited quite a lot of folks in and out of the Agency is a responsible way to use taxpayers’ hard earned money.

On a personal note, I was brought in, a political appointee, to head the new Office of Vision and Strategy. It was my duty to do precisely what I did. My fifteen months at SSA were wonderful and I was honored and privileged to do public service. I witnessed the passion of so many employees who were dedicated to their mission and worked so hard to get the job done. I feel that I owe it to them, and to the American citizens, to continue to educate people about the challenges and opportunities that Social Security has moving forward.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s