Government Reform

In late December I spent the night with 60 homeless folks, sleeping in the church where my wife and I were married. I do it every year. My job was to keep awake. I was there over a Saturday night.

One of the homeless guys got up just before midnight, and started trying to call his unemployment office. He told me that it was HIS job to call in, every few weeks, to tell ‘em that he was, yes, still unemployed. That’d keep his checks coming. No call, no checks. Of course, where and when there’s a lot of unemployment, it’s almost impossible to get through.

He’d found out that the BEST time to call is right after midnight, on Sunday morning. There are operators “always on duty,” but there are far fewer people calling right about then. After many calls, he got through just before 1:00AM.

I had to wonder, how many Congresspersons, political appointees, or even civil servants ever thought that homeless people would be calling the unemployment offices just after midnight on Sunday. I can’t believe it’s intentional.

Couldn’t they hire a few of the unemployed folks to man those phone banks to take the calls of the OTHER unemployed? Shouldn’t they be thinking about how their ’systems’ are used? Is that “rocket surgery”?

Turns out that “usability testing” for websites, etc., is a recognized trade, and one of its gurus, Steve Krug, (among others) has written two books about it  – one of them called, appropriately enough, Rocket Surgery Made Easy. There’s an association for that, and even a journal!

Does government ever do “usability testing” on its programs? When I was a Deputy Director of Al Gore’s National Performance Review, we were focused on the citizen as “customer,” but I don’t remember anything quite like “usability testing.” It would have been a good idea!

…it’s not too late.

I heartily recommend reading Atul Gawande’s new book, “The Checkbook Manifesto.” I’m not alone…

In it, Dr. Gawande recounts the birth of the modern formal checklist, developed by the US Army Air Corps (now the Air Force) after the crash of a new four engine Boeing bomber on a test flight in 1935. Flying a four engine plane was so much more complicated than flying a two engine plane that the pilot, the Air Corps’ chief of flight testing, left something out. Like releasing a locking mechanism on the elevator and rudder controls…

Dr. Gawande ascribes the safe landing of US Airways Flight 1549 on the Hudson to the crew’s following their checklists. (BTW, Captain “Sully” Sullenberger does too.) The “hero” was the checklist!

He goes on to describe the checklists used by construction companies in building our enormous shopping centers, hospitals, and the like. No longer do they rely on “Master Builders” keeping everything straight in their heads.

He recounts a study in which Intensive Care Unit patients required an average of 178 individual actions per day, and by and large the doctors and nurses missed only about 1 percent. But that’s two a day…

He then describes in detail how he and a team developed surgical checklists for the World Health Organization, testing them in a handful of hospitals of varying sizes in countries ranging from Canada to Tanzania. The health benefits were enormous across all hospitals.

Dr. Gawande points out that the senior surgeons generally rejected the idea that THEY needed to follow checklists – why, they’d been trained by experts and had years of experience! Yet, shown the results, they admitted that if THEY were the patients, they’d like their surgeons to use checklists, certainly.

The book is hard to put down, and I’m about to write up two checklists showing what to do at our house when we lose internet access. I already use a checklist for closing up our house in West Virginia – but it needs updating. I am sold!

The book has lessons for the design of government as well. Here are the last two paragraphs of his introduction (page 13):

“Here then is our situation at the start of the twenty-first century. We have accumulated stupendous know-how. We have put it in the hands of some of the most highly trained, highly skilled, and hardworking people in our society. And, with it, they have indeed accomplished extraordinary things. Nonetheless, that know-how is often unmanageable. Avoidable errors are common and persistent, not to mention demoralizing and frustrating, across many fields – from medicine to finance, business to government. And the reason is increasingly evident: the volume and complexity of what we know has exceeded our individual ability to deliver its benefits correctly, safely, or reliably. Knowledge has both saved us and burdened us.

“That means we need a different strategy for overcoming failure; one that builds on experience and takes advantage of the knowledge people have but somehow also makes up for our inevitable human inadequacies. And there is such a strategy – though it will seem almost ridiculous in its simplicity, maybe even crazy to those of us who have spent years carefully developing ever more advanced skills and technologies.

“It is a checklist.”

So if complexity is a part of our problem in designing governmental interventions that work, then Dr. Atul Gawande would argue that checklists are part of the solution.

But it is here, as he winds up the book, that he makes the point most relevant to the design of government (page 185):

“We have a thirty-billion-dollar-a-year National Institutes of Health, which has been a remarkable powerhouse of medical discoveries. But we have no National Institute of Health Systems Innovation alongside it studying how best to incorporate these discoveries into daily practice – no NTSB equivalent swooping in to study failures the way crash investigators do, no Boeing mapping out the checklists, no agency tracking the month-to-month results.

“The same can be said in numerous other fields. We don’t study routine failures in teaching, in law, in government programs, in the financial industry, or elsewhere. We don’t look for the patterns of our recurrent mistakes or devise and refine potential solutions for them.”

Sounds like he also thinks that the lack of feedback mechanisms is also a problem, doesn’t it?

We can no longer afford the Congressional legislative process described some years ago by then Indiana Congressman (and later President of New York University) John Brademas: “Congress never gets anything right the first time – after five or six years, we have to revisit our ‘solutions’ and correct them.”

In an era of increasingly complex, critical, and urgent problems, that’s simply UNSAT.

I’m starting to think about checklists as part of any “Legislative Impact Statements.” Not rocket surgery, is it?

Check out Glimmer by Warren Berger.  I’ve been looking for a guide to the world of design, and Glimmer, with a wonderful website, is a whole sack of Sacagaweas.  I’m not to the bottom of it yet.

[BTW, I’m recusing myself from putting “If We Can Put a Man on the Moon…” by Bill Eggers and John O’Leary at the top of the list, for two reasons. First, I know both authors, and second, I’m listed there under Acknowledgements. I’m forever grateful for their referencing Improving Program Design, a publication of Vice President Al Gore’s National Performance Review that I both inspired and demanded.]

I WOULD like to offer a tip of the hat to Ralph Caplan,  a pioneer of modern design, for the following paragraphs from his book By Design:

“But aren’t the products of design hair dryers, computer screens, cereal boxes, curtains, bedspreads, vacuum cleaners, kitchen appliances…

“Sure. Also chairs and computer programs and office partitions, space capsules and tractors, restaurants and stores and cities, films and books, and government legislation and protest strategies.” [Emphasis added]

That was the first reference I’ve seen to ‘designing government’ since I first wrote “The Design of Government” early in the Carter Administration. It’s about time!

And only this year I found the second reference to ‘designing government’ — in a blog, Design Thinking, by Tim Brown. He’s one of Warren Berger’s glimmerati! I commented on his post on “Redesigning California” and you can see my entire comment there. My key sentence is:

“The need is for (a) design thinking, (b) familiarity with the known Tools of Government, and a deep understanding of the sciences of complexity (from cybernetics to chaos).”

Glimmer’s website has a hyperlinked list of the Glimmerati. What more do you need?

…world enough and time…

If you asked your 16 year old son to clean up the garage, but told him that he’d get no dinner until he’d cleaned up his room, would you be surprised the next day to find the garage as messy as ever? Not hardly.

Now suppose you were Daddy Hank (Paulson) or Uncle Tim (Geithner) passing out billions to the biggest American banks. You tell them you’d really like them to resume lending, and get America on its way out of the Great Recession. But you also make it clear that they won’t get to set their own salaries (and bonuses!) until they repay the billions you’re lending them. Would you be surprised to find that they’re paying back the loans without doing any appreciable lending? Not hardly.

So — recently, the Bank of America paid it all back so they could make “a decent offer” to someone to come in and be the boss. And now Citibank is about to repay its loan. And are they lending to help homeowners and small businesses weather the recession? Not so much.

(As an aside, according to the Times article cited above, Standard and Poor’s “wrote a remarkably candid research note that suggested the $45 billion repayment didn’t really matter, because if the bank got in trouble again, taxpayers would be there with another bailout.” That’s probably true. Cf “moral hazard.”)

And today’s news is that Grandpa Obama is taking the bankers to the woodshed because they’re not lending. Do you really think that this will be effective? You’re alone…

Some guy writing under the pseudonym “Smith” wrote the last word on this some years ago: “It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own self interest. We address ourselves, not to their humanity but to their self-love, and never talk to them of our own necessities, but of their advantages.” – Wealth of Nations, Book 1, Chapter 2.

So going back to your 16 year old son for a moment. Suppose you told him that he could drive your old pickup to school if (and only if) he dropped off his younger sister at school – each day. Or if you said, if your soccer team wins State, you’d give him the car? Do you think that’d change his behavior on an ongoing basis? At least until that championship game? Have you ever been a 16 year old that wanted a car?

In the Federal government’s Executive Branch there’s been an emphasis on identifying and measuring “inputs, outputs, and outcomes” for some years. Using this paradigm, one can discuss how the Feds might have leveraged the big bankers into helping others besides themselves.

First, the “inputs” are (a) the bailout money, (b) the ongoing efforts of the banks’ staff, and (c) the ingenuity of the banks’ management. No surprises there.

Second, some useful “outputs” might have been the leveraging of the bailout monies into useful loans to small businesses and homeowners. We might have said to Citibank: “You won’t get the right to set your own salaries and bonuses unless and until you show us that you’ve put that bailout money to work!” And there are metrics for that.

Third, we might have focused instead on “outcomes.” We might have said to the big bank bosses assembled, “Since we just saved your ass(ets), we will control your salaries and bonuses until the unemployment rate goes down to, say, 7%. YOU go figure out how to make that happen.”

And then the interests of the big banks and the government and America itself would have been aligned, which is always useful. And since the big bank bosses are the thought-leaders of their industry, they could have influenced the behavior of all the lesser banks as well.

And yeah, I know it’s probably not quite that simple. Maybe we’d have had to say, “Reduce unemployment at a steady rate through actions that don’t increase inflation beyond such-and-such.” Given that the economists working for the banks are not THAT much smarter than academic economists (although my guess is that industry modeling is better than academic modeling), I think that we could have defined “where the ditches are” with adequate accuracy.

So the TARP legislation was written in such a way that a 16 year old could have gamed it, much less the Masters of the Universe. And that leads to a question: Didn’t the Paulson team (and it’s inheritors) REALIZE how easily it would be gamed? Or was the structure of the bailout “suggested” by the industry itself, and not considered thoroughly? Or did the Bush Treasury team really think that the big banks would have refused a more stringent plan? Perhaps we’ll never know; there’s not that much transparency in government!

The underlying premise of this blog is that insufficient consideration of (a) unintended consequences and therefore (b) alternatives are at the heart of such governmental failures. After all, it was John Brademus, then a Congressman from Indiana and now retired president of New York University, who said, “Congress never gets anything right the first time. After five or six years, we have to revisit our solutions and correct them.”

Unfortunately, we are living in a time where the problems are both complex and urgent, and the solutions can’t always be ‘revisited’ and corrected later. Try health care, much less global warming.

(BTW, I refuse to believe that the big banks had such a stranglehold on our government that they could dictate the terms of legislation that was so easily gamed. That may well be true, but if so it signals the transition from Republic to Empire that brought on the Dark Ages when it happened in Rome. That is, when private interest trumped public interest without question. I’m just not going there.)

So if we need more attention to how government interventions in society’s outcomes and in the workings of the private sector (Note: while under TARP, the big banks were part of the “semi-private” sector, which includes the defense industry. GM is still there…), then we need better “design.” There are at least three ways to get there.

First, we need more attention to the formal world of “design,’ as for example taught at “the d-School” at Stamford. More on that in upcoming posts.

Second, we might go with “Legislative Impact Statements,” as now required in Australia and New Zealand. See Bob Zarnetske’s comment (#3) to my posting entitled “Legislative Impact Statements.” This would require some sort of institution, itself subject to manipulation. It’s worth considering, though.

Third, we might go beyond the “notice and comment” procedures now used in the Federal regulatory process and go directly to wide-open web-wide commenting. Massive collaboration seems the wave of the future anyway. This, too, would require real people to sort and review the comments, but they would be under widespread scrutiny as they did so.

There’s a fourth way: All of the Above.

Today’s Washington Post column by Allan Sloan includes the following:

– “Cash for Clunkers.” It was a well-intentioned plan that was supposed to increase consumer confidence, spur fuel efficiency, jump-start the auto industry and help create American jobs. Instead, it disproportionately benefited foreign automakers, which create fewer North American jobs per car dollar than the Detroit Three do. And sales came mostly from inventory, doing little to increase production and jobs. What’s more, by junking clunkers, the program removed many low-end vehicles from the used-car market, running up prices for the lower-income people who’d normally buy them. So we hurt the people most in need of help, while throwing taxpayer dollars down the drain. As the saying goes, the road to hell is paved with good intentions.

It’s easy to throw up your hands when talking about “unintended consequences,’ but none of these can really be said to be “unforeseeable consequences.” In hindsight they look pretty obvious to me! And I believe that a little foresight would have revealed them as well.

We already have Environmental Impact Statements; they’re either famous or infamous, depending on your viewpoint. There are also financial impact statements: how much will this bill cost, or this revenue measure raise?

Maybe we need to have the Congress publish “Legislative Impact Statements” that list what the legislators think will happen, both good and bad, so that we can judge afterwards how much they thought about what they were doing?

If you doubt the role of chaos in this world, read this Washington Post story about how the Berlin wall fell.

Talk about a butterfly’s wings causing a tornado!

A decent provision for the poor is the true test of civilization.

—Samuel Johnson, lexicographer (1709-1784)

_____

November Fifth is known in Great Britain as “Guy Fawkes Day,” named in honor of the man who was caught trying to execute a plot to blow up the English Parliament in 1605.

American legislators sometimes neglect to do things so blindingly obvious that Guy Fawkes almost becomes a sympathetic figure!

This year’s candidate for the first Annual Guy Fawkes Award for Legislative Obtuseness (the “AGFALO”) goes to the Maryland Legislature, which just now is considering increasing Maryland’s state guidelines for child support payments – for the first time in twenty years!

According to the story in the Washington Post, Maryland (a) has the highest per capita income in the United States, (b) is forty-first in what parents pay for child support, and (c) obviously has no sense of shame). This affects half a million children in the state. Half a million!

The good news is that they’re considering raising the support payment guidelines; the bad news is that the legislation also contains a provision lowering the support payments for lower income parents!

But the issue is not whether Maryland should raise the payment levels, or should have done so in 2007 when the District of Columbia did. Or whether Virginia “does a better job” since theirs were raised in 1995.

The issue is far simpler than that: Should the vagaries of the legislative calendar and the shifting winds of politics determine whether kids in broken homes get fed? Viewed in this light, it becomes a design of government issue!

What does Maryland have that’s more precious than its children? How will Maryland take care of people who grow up in homes that can’t afford to feed and clothe them, or help them focus on education for the jobs of tomorrow?

Wouldn’t it also help keep families together, and keep the kids on track, if the parents knew in advance  that they faced stiff child support payments?  Those payments must have been a joke for the past ten years!

Now, what could be simpler than INDEXING the guidelines to the Consumer Price Index?  Or some other index that accounts for inflation? Is this so hard? It seems to work well for Social Security, and I bet it works well for the pensions of Maryland’s legislators and staff!

Given that the US Congress has somehow neglected to index the minimum wage, even though some foreign countries and even some of the states have done so, I sometimes wonder. (see this blog, page 33)

Are all legislators both (a) so comfortable within their work lives and (b) so worried about possibly getting divorced that they can’t see the damage they are doing to our image of effective government, much less the harm that’s done to poor individuals and the future of America?

For he’s a jolly AGFALO; For he’s a jolly AGFALO; For he’s a jolly AGFALO — Which nobody can deny!

_________

[BTW, enforcing the child support payments is also a major problem that is not addressed here, nor is it addressed well in Maryland or elsewhere.]

_________

ADDENDUM: I attended a public meeting about the proposed child support guideline revisions, and was subsequently quoted in the Annapolis Capitol newspaper.

You may know that the Washington Post is holding a contest for their next Post Pundit. Here are the first round winners.

Yes, I applied. No, I wasn’t chosen. Of course, I’m a “one-trick pony,” so I wasn’t too surprised. And if you’re reading this, you already know the trick…

They asked for a 400 word essay and a 100 word statement of why you should be chosen. See below. I’ve put links into the essay, and expanded the statement slightly.

-=-=-=-=-

How Well Is Our Government Designed?

Did Congressman Elton Gallegly (D, CA) ever think that his “Crush Video” law, protecting kittens and mice, would cause the stampede of elephants to the Supreme Court in United States v. Stevens? The elephants, from the New York Times and National Public Radio to the National Rifle Association, have paid lawyers handsomely. Would opening draft legislation to public comment via the Internet help prevent such unintentional consequences?

The government spent this spring “adjusting” the analog-to-digital TV transition process, addressing delays, communication problems, underfunding, and finally oversubscription. The program had trouble reaching poor, elderly, rural, and non-English-speaking citizens – those who really need emergency warnings. Did anyone know about Everett Rogers’ work on the communication of innovations?

The small staff at Commerce’s National Telecommunications and Information Administration (NTIA) gives tiny grants and focuses on policy and research. No surprise that the surge in coupon applications “crashed” their program. And so no surprise when the “Cash for Clunkers” surge crashed at another small agency, Transportation’s National Highway Traffic Safety Administration (NHTSA).

Memo to the Congress: Don’t give any shipbuilding contracts to Seven-Eleven.

The computer model for costing health insurance proposals at the Congressional Budget Office gets no third-party review of its assumptions or internal workings, according to a story in the Washington Post. CBO’s report to Chairman Baucus on its latest results has no confidence intervals, either. “Your mileage may vary,” as Phil Ellis says, but CBO is not estimating by how much.

Has the health insurance industry got better computer models that can second- (or third-) guess CBO’s? The better their models, the better their negotiating strategies.

Phil Angelides’ Financial Crisis Inquiry Commission faces far more difficult challenges than the Pecora Commission of 1933.  Computers and telecommunications have made the world’s financial system far more complex and more interrelated. And the financial industry has long been able to afford excellent computer modeling. If the Angelides Commission is to dredge its recommendations for loopholes and unintended consequences, it must have access to even better models. Otherwise, the industry will out-game the Commission every time.

Keeping the global financial system “between the ditches” throughout this century will require more than modeling. The Commission needs expertise in several post-WWII disciplines, including game theory, chaos theory, and complexity itself.

‘Intelligent design’ deserves a seat at the table in Washington. The private sector is ahead of us again. See Business Week’s Special Report on Design Thinking.

As Jimmy Carter learned to ask, “Why not the best?”

-=-=-=-=-=-

My unique policy perspective: Thirty years a Fed, nineteen as a Senior Executive. Served in seven Cabinet departments, five agencies, and the White House under both Ford and Clinton.

I first wrote about ‘designing government’ during the Carter Administration in a paper now on my blog.

As a Deputy Director of Al Gore’s National Performance Review, I learned that private sector innovation must be continually infused into government.

Sadly, the Federal government only looked to the private sector for inspiration in an organized way four times during the 20^th century: the Brownlow Committee (1937) the Hoover Commissions (1947 & 1953), the Grace Commission in 1982, and Vice President Gore’s National Performance Review (1993).

While with Gore I oversaw another report on program design. Same problems, twenty years later.

Harvard’62, Georgetown Law’72

Who’s Who in America (first listed in 1984)

Kennedy School’s Innovation Awards Program judge

Full resume here.

The $8,000 First-time Homebuyer Credit program has been riddled with fraud and abuse, in ways the IRS didn’t foresee or try to control. Here’s the story, and a handful of ways they could have sought out such “unintended consequences.”

A recent Washington Post story outlines how hundreds of millions may have already been paid out to people who fraudulently or mistakenly took advantage of the first-time homebuyers’ $8,000 tax credit funded under the American Recovery and Reinvestment Act of 2009:
• 19,300 people claimed $139m on their 2008 tax returns
• 74,000 people claimed nearly $500m for unqualifying purchases
• 580 people under 18 – including 4 year olds – claimed $4m

Of course this is small potatoes – perhaps 1.4m households have claimed almost $10B in tax credits. And the Congress is rushing to extend the tax credit provision, proving again that “fraud, waste, and abuse” are in the eye of the beholder.

But the IRS has identified 160 potential tax credit schemes and selected 107,000 claims for reexamination. So they’re looking at 7.6% of those claiming the credit! What a waste of resources, if these problems could have been headed off at the pass!

[NB: Our concern here is the mechanics of program design, not (a) whether the program was a good idea, or (b) whether it has jumpstarted the housing market.]

The Post story and the House hearing it reported on were based on a report by the Treasury Department’s Inspector General for Tax Administration. The Inspector General asked the usual questions: (a) What happened, and (b) What can be done about it?

The first page of the IG report notes that “the President’s mandate with regard to stimulus payments is to prevent fraud, and not simply minimize it or address it after it has occurred.” Prevention is a hallmark of Inspectors General; it even appears on their website, sponsored by the unfortunately named Council of the Inspectors General on Integrity and Efficiency (CIGIE calls up “ciggies” — where was The Acronym Control Team?}

It was only 100 years ago that philosopher George Santayana said, “Those who cannot remember the past are condemned to repeat it.” We also need to understand the past. This leads to two more questions: (c) How did it happen, and (d) How can we prevent similar problems in the future?

The problem, simply stated, is that Treasury is offering substantial sums to countless potential takers without finding ways either (a) to prevent fraud or error, (b) to discourage it, or (c) to make it easily discoverable after the fact.

This has had the effect of increasing both fraud and abuse, and also increasing substantially the efforts that the IRS must undertake subsequent to the claims being filed. It is worth noting that Ashby’s Law clearly applies here: the taxpayers outnumber the IRS, so prevention will be much, much easier than remediation.

The IRS did develop a special form (Form 5405) to catch claims in excess of that allowed and claims by those with adjusted gross incomes above the set limits, as well as claims without Form 5405 attached (Duh!). The IRS, however, did not use the Form 5405 to verify eligibility and no proof of a home purchase was required, such as the ubiquitous HUD-1. Both had been recommended by the Inspector General.

So that’s WHAT happened. But HOW did it happen? As it is, we have no idea whether anyone tried to think through the likely results of their program design, or cast about looking for unintended consequences. Appendix I of the report is silent on any inquiry into the thought processes of the IRS staff who designed the tax credit program, and unless the redacted portion of Page 5 hides some names, we have no idea who was involved in the decisions.

Of course, the Inspector General could assign pseudonyms to the actors in this horror story; perhaps “Mary Shelley” as an Assistant Commissioner, “Edgar Poe” as a career staffer, “Sarah Langan” as the staff director of the relevant congressional committee, and “Bram Stoker” as an Associate General Counsel. Then their deliberations could be revealed while safeguarding their identities.

But if anyone HAD BEEN looking for ways to search out unintended consequences, here are a few ways they could have done so, ranging from the obvious to the quite innovative:

1. Ask those directly involved to think of possible unintended consequences over the weekend and come to a meeting on Monday to address that subject only.

2. Assign a small staff group to collect and present possible problems in a meeting with program advocates, with the political appointee acting as “referee,” not as advocate for any one view. This is a variant of “redteaming,” as practiced at the Pentagon or even in theory at the Federal Aviation Administration (example).

3. Ask staff at “the point of the lance” what could go wrong. Field staff have a different perspective on what can happen in “practice” versus in the “theory” as developed by headquarters.

4. Assemble the Senior Executives from across Treasury who have received Distinguished Rank Awards. (See note) This group has more years in the pay line than the immediate office has in the chow line, and can bring that experience to the problem at hand.

5. Email a description of the proposed program for comment to a handful of members of the relevant “policy communities” around Washington, whether the Brookings/Cato/AEI group or the Booz/IBM/Deloitte group, or both.

6. Post the description on an open internet site, asking for insights from the world at large. Let anyone comment, or even become Facebook “fans” of Tax-Breaks-For-First-Time-Homeowners.

In all likelihood, “None of the above” was the option chosen.

BTW, my preferred solution would have been (a) to have the buyers execute a form before a Notary Public swearing to the facts entitling them to the tax credit, and (b) to have that form registered along with the new deed and mortgage at the county land office.

I’d guess that far better solutions would have arisen from almost any of the methods outlined above.

PS: A friend suggests that the IRS check the claimant’s recent filings to see if mortgage deductions were taken during past years. That would certainly help!

-=-=-=-=-

Note: Distinguished Rank Awards. There are about two million career civil servants. About six thousand of these (0.3%) have been promoted, based on merit, to become members of the Senior Executive Service. About sixty members of the SES (thus only 0.003% of the two million) are given Distinguished Rank Awards each year, based on nominations by their departments and agencies. Few groups know more about the federal government, and few are less often asked for their opinions of how to improve it.

Disclaimer: I received one of the awards in 1996.