The Cup Runneth Over

Try this experiment in your office. Take a cup and fill it to the top with water. Now, take another cup and fill it half-full. Pour the half-full cup into the full cup without spilling any water. How'd it go? This is exactly what is happening nationwide with public assistance - food stamps, unemployment, housing, utility assistance and more. The economic downturn has increased demands on these systems by 40 percent to 50 percent, and the impact has been devastating. The water (customers) is spilling everywhere, the cups (dedicated case workers) are breaking and everyone is freaking out. What is important to understand is that the economic crisis didn't cause these system problems - it amplified them. To be more precise, it has amplified three "best practices" that simply don't work:

Automate the cups. Two initiatives have been particularly popular in public assistance. The first is to spend $100 million on a case management system so we can keep track of all the water. The second initiative is to put the benefits application online so that the water gets to the already full cups sooner.

Outsource the cups. Clearly private-sector cups hold more water. Except they don't - just ask the states who have to "insource" their failed "outsourcing" efforts. Outsourcing doesn't give you bigger cups, nor better cups - they are just a different set of cups with a horrible contract and incentives perfectly aligned to encourage more water - not less.

Hold the cups accountable. Clearly the cups aren't working hard enough so we need to create some performance standards and punish the cups for not meeting them.

Put this way, these solutions all sound absurd. And they are. Not only for public assistance, but all government systems. The problem we are facing in government is not a technology problem or an accountability problem, and it certainly isn't because we don't wear corporate logo shirts. As I explained in detail in my column "Free the Hostages," private-sector monopolies are no better than public-sector ones. Our problem in government is simple and universal - capacity. We don't have the capacity to do all the good that is required.

How do we increase capacity? We only have two choices: Increase the size of the cups, or decrease the amount of water. And in today's budget climate we really only have one choice because we aren't going to get bigger cups. That's okay, the real problem is the water anyway.

When you stand in the lobby of a public assistance office, it is literally like standing in a pool of rising water. It's beyond grabbing a mop or plugging a hole. There is a tidal wave of people all waiting to wait; taking a number to take a number.

When we study the reasons people call or visit the office, however, something glaring jumps out. Less than 30 percent of the people are there to get benefits. What are the other 70 percent doing?

• Forty percent of them are simply trying to figure out what is taking so long. The great UK management thinker, John Seddon, calls this "progress chasing". I call it the "Crazy Cycle", as per a column I wrote in September 2008. Progress chasing includes:
• "I sent in my application and want to make sure you got it."
• "I haven't heard back from you and wanted to see where we are in the process."
• "I left you a ton of messages and no one returns my calls."
• "Maybe you lost it so I resent it."
• "I was afraid you would lose it so I brought it in."
• "Where is my stuff?"
• "Seriously, where's my stuff?"
• The customers trying to figure out where they are in this clogged-up system are the very ones clogging the system in the first place. The more progress chasers call, the more employees are required to handle the calls; which means they get less work done, fall further behind and get more calls. Eventually we wake up five years later and find that the phone center staff is now as big as the processing staff and the office is spending $50 million on software that only tells them how far behind they are.
• Seventy percent of customers can't complete their transaction on the first visit. Whether applications are too difficult, policies are too restrictive or communication is unclear, customers come in the door confused and leave more confused. While the office gets credit for "serving them," those customers will clog up the system trying to find out what they are supposed to bring, who they bring it to and why the office lost it last time.
• Many programs require a renewal process - which is usually less cumbersome for clients and staff - if they ever get to it. Offices nationwide are so far behind they can't keep up with annual renewals. Typically 50 percent of clients don't get renewed on time, which means they must return to the office to go through the arduous initial application all over again.

These are all self-inflicted wounds, stemming from one problem: we are too slow. Is it a worker problem? Spend a day in a public-assistance office and you will never say that again. Public-assistance workers, like most government workers, are hard working, mission-driven, dedicated people. But they are stuck in complex, rigid, soul-crushing systems. One of the questions we ask caseworkers at the Change and Innovation Agency in our system redesign efforts is: When was the last time you took a day off without feeling guilty? More often than not, the answers are coated in tears. Our research shows that in a typical day a caseworker only gets to spend 50 percent of their time conducting interviews and determining eligibility. Government workers are not slow, they are behind.

In the case of public assistance we are behind because we have contracted a not-so-rare condition called Parkinson's. I am not referring to the ailment that Michael J. Fox has brought so much attention to. Rather this condition derives from Parkinson's Law: Work expands to fill the amount of time allotted for it. If you allot 30 days to get something done, it will take 30 days. If you give the same chore 60 days to be completed it will take 60 days. The condition also has another name: the gravitational pull of legislative standards. I've seen this repeatedly in permitting processes where 95 percent of permits are processed in 90 days, but a few outliers take a really long time. Those outliers complain (rightly so) to their elected officials who then mandate that no permit can take longer than 180 days. Even with the same workload, same processes and same staff, guess how long all the permits will now take? 180 days.

Inside a typical public-assistance process is two hours of hard, value-added work per customer. Various legislative and federal mandates have created 30-day timeframes to complete these two hours of work. Guess how long each customer ultimately takes? This is not because employees aren't working hard, but because their cups are filled with 30 days' worth of water at a time.

Parkinson's by itself can be debilitating, but it usually attacks our system in conjunction with a second malady that Eli Goldratt calls the "student syndrome". Simply put, we do our homework or study for the test the night before it's due; which leads to stress, sleep deprivation and caffeine jitters. The student syndrome in a public-assistance office is equally unhealthy as staffers stress over how they will address the 100-plus cases due that day, each of which requires two hours of work.

In short, the longer something takes, the more work it creates and the more it costs.

What happens when water becomes stagnant? Bacteria grow, everything stinks and we get sick. The same is true when customers stop flowing through our systems. The backlog grows, everything stinks and our organizations get sick, to say nothing of the effects on the customers &151; a family can't get food for 30 days, a job for 60 days or housing for 6 months.

What's the solution? How do we stop all this madness? Do two hours' worth of work in two hours. If we do two hours' worth of work in two hours we don't get behind, we don't create extra work and we increase capacity. Customer service and customer satisfaction improve, stress abates and caseworkers are able to spend more time doing what they joined government to do: help people.

The results speak for themselves. In states that have implemented this solution - driving down the 70 percent extra work that is self-inflicted, speeding up the process and allowing caseworkers to focus on value-added tasks - the results have been as follows:

• 40 percent more customers served
• 70 percent faster
• 50 percent fewer errors
• 0 new staff and
• 0 dollars of technology investment

The water has receded, the pressure is off and now it's the customers' cups that runneth over - with gratitude.