Kenneth Zawodny, OPM's associate director of retirement services, said he hopes a series of incremental IT upgrades will speed up processing of retirees' pension claims. (Thomas Brown / Staff)
By 2016 or 2017, the Office of Personnel Management hopes to have a major IT upgrade in place to speed up the processing of federal retirees’ pensions.
But even when the new Case Management System, or CMS, is in place, the government will still be decades away from its ultimate goal: a completely automated system that calculates new pensions with minimal human involvement.
OPM has struggled for two decades to fix its sluggish pension processing system, and wasted $100 million on four failed attempts at information technology systems it hoped would do the job. The most recent failure was an ambitious, end-to-end system called RetireEZ, which OPM canceled in 2008 after it flunked numerous tests.
Meanwhile, retirees often wait months — in some cases, more than a year — for their full pensions and in the meantime must get by on partial annuities that are sometimes far less than what they are owed.
This time, OPM is planning a more modest system of incremental IT upgrades. And Ken Zawodny, OPM’s associate director of retirement services, said in a May 22 interview with Federal Times that he hopes taking the more conservative approach will yield success — though OPM said it is too early to predict exactly how much processing time or accuracy might improve when the upgrades are done.
“Our Case Management System is basically the engine in the entire process,” Zawodny said. “There are many pieces that are going to tie all of our activities together into the workflow of retirement actions and operations.”
One of the main problems preventing a completely automated system is the fact that the personnel information needed to calculate pensions largely exists in two states: structured data that is entered directly into a computer and can easily be manipulated digitally, and data found on paper records that have mostly been scanned into PDF form.
Since 2005, the government has been keeping federal employees’ personnel information in the form of completely digital structured data. Employees hired since that year, when they decide to retire, will have a career’s worth of easily readable digital data recording their work history, making it considerably easier and faster to calculate their pension benefit.
But because employees likely to retire soon were hired well before 2005, they have years or decades of service recorded in PDF form. And those records require OPM adjudicators to open them up on their computer and eyeball them to find the information needed to calculate their pensions. That includes whether they are under the Federal Employees Retirement System or the Civil Service Retirement System, when they joined the government, if and to what extent they have served in the military, what they have been paid and many other pieces of information.
The first part of OPM’s upgrade, which it began rolling out last year, seeks to make the viewing of scanned documents easier. The Web-based Data Viewer allows OPM’s adjudicators to look at service records and other retirement documents scanned in PDF form, and it will help agencies assemble more accurate retirement packages before they are sent to OPM.
One of the major problems with the pension calculation efforts is that agencies frequently submit retirement packages with incomplete or inaccurate information. This forces OPM officials to waste time tracking down or correcting those documents before they can start calculating a pension. OPM hopes Data Viewer will help solve that problem. Data Viewer is now being piloted at 11 agencies, including the Veterans Affairs and Homeland Security departments and payroll providers at the Interior and Agriculture departments, and Zawodny said he expects it will be used throughout the government within 12 months.
The next phase of CMS would begin in fiscal 2014, assuming Congress authorizes OPM to spend $2.6 million from the Civil Service Retirement and Disability Fund on the project. If all goes well, Zawodny hopes the project would be done in two or three years.
Zawodny said CMS will let OPM pull the post-2005 structured data from other agencies’ computer systems and enter that data into a calculator that — when relevant data from PDFs are also entered — will crunch the necessary numbers to find the employees’ pensions.
CMS also will let OPM review case files submitted from other agencies to make sure no vital records have been forgotten and find any gaps in service or discrepancies that need to be resolved. And OPM will be able to better track and report on retirement documents and cases, Zawodny said.
OPM said in April that it is now processing claims in roughly 77 days, half of the 156-day average in 2011. Since the beginning of 2012, OPM has hired and trained more staff to process retirement claims and answer questions from retirees, greatly increased its use of overtime to process more claims and streamlined some of its processes to get that claims time down. Zawodny hopes CMS will help OPM cut that time even further, but could not say by how much.
Zawodny said OPM is trying to decide whether it will do the project in-house or contract out some of the work. He could not estimate how much the entire project will cost, but said he does not expect to spend as much as $2.6 million each year, meaning it would probably cost less than $7.8 million. OPM spent about $21 million on RetireEZ before pulling the plug.
But the dream of an almost entirely automated system that relies on structured data will likely not be a reality for decades — not until the employees who joined the government beginning in 2005 start to retire.
“Until the folks who have gone through their entire life working as a federal employee and retire, when all their data has been collected and stored electronically as data elements, only then will we be able to realize the full potential of what we’ll be able to do,” Zawodny said.
OPM doesn’t have the manpower to go back and key in the relevant data from those millions of PDF records, Zawodny said, and doing so wouldn’t be cost-effective.
OPM is also challenged by the fact that the dozens of independent agencies often have their own record-keeping, payroll and human resources systems. “Getting them all to sync up at the same time isn’t going to happen,” Zawodny said. “It’s going to be over a long time period.”
The National Active and Retired Federal Employees Association said OPM’s incremental strategy is probably the best way forward. It’s a more realistic approach that takes budget limits into account, NARFE said, but it will not result in radically quicker pensions.
“The end goal here is to be able to have that data, as an employee goes through a federal career, automatically put into the system that is ready to crunch when he or she is finished, and spit out an annuity,” said Dave Snell, NARFE’s retirement benefits director.
NARFE legislative director Jess Klement said her organization is concerned by OPM’s decision to take the $2.6 million from the retirement trust fund and not to ask Congress for new appropriations. “That sets a very dangerous precedent,” Klement said.
OPM said that it would not be the first time it has used money from the trust fund to pay for technology upgrades. The trust fund was used to pay for the aborted RetireEZ system.
All expenses that OPM incurs in administering the retirement programs are transferred from the trust fund, including IT enhancements, OPM said. Congress establishes a limit on the funds that OPM can transfer from the trust fund for administration of the retirement programs in the annual appropriations bills. OPM’s fiscal 2014 budget request seeks a limitation amount that includes the $2.6 million for IT enhancements. OPM said it is seeking to retain access to those funds for more than a year.