Memo to Musk & Ramaswamy: Incentives Count!

As co-leaders of a “Department of Government Efficiency,” Elon Musk and Vivek Ramaswamy are tasked by President-elect Trump with finding ways to “dismantle Government Bureaucracy, slash excess regulations, cut wasteful expenditures, and restructure Federal Agencies.” Judging from recent tweets from Musk and others, this effort will be all about finding and cutting government waste.

We’ve been down this road before and it leads to a dead end. Back in 1993, less than two weeks after Bill Clinton took office, Penguin books released Reinventing Government, a book that promised to “transform the public sector.” Clinton was so enamored with the ideas in the book that he quickly created a National Partnership for Reinventing Government whose goal was to make a government that “that works better and costs less.” Clinton placed his vice-president, Al Gore, in charge of the “partnership.”

Over the next several months, Gore held numerous town hall meetings seeking ideas from the public on how to reinvent government. In September, just six months after the partnership was created, he issued a 175-page report that contained 384 general recommendations and 1,250 specific proposals aimed at improving dozens of government agencies.

There was certainly room for improvement. A few years before, a Forest Service staff officer admitted to me that, when he took his current job, he realized that only nine of the seventeen people working under him were actually doing anything. Almost half of his staff were just pushing paper around. However, he decided not to do anything about it because if he only had nine people working for him, his pay grade would be reduced.

The original book, Reinventing Government, focused mostly on the “works better” part of Clinton’s mandate. Using ideas such as letting government agencies keep any budget surpluses they had at the end of the year, the book proposed to create an “entrepreneurial spirit” within the bureaucracy. Of course, bureaucracies are already entrepreneurial, but their aim is generally to get and spend more tax dollars, not provide government services more efficiently, which is what the book sought.

In the end, however, Gore decided that making government work better was just too complicated and might not produce results in time for the next election. So the actions the administration took after the report came out focused on making government cost less. Most importantly, Clinton and Gore ordered most government agencies to reduce their work force and spending by 10 percent. This number was entirely arbitrary, but it allowed Clinton and Gore to brag that they had eliminated 250,000 federal jobs, saving taxpayers $40 billion a year.

Some of those gains were from the reduction in defense spending due to the ending of the soviet threat, but some of them were real. The problem is they were only temporary.

One person’s waste is another person’s livelihood. In general, government patronage works by concentrating benefits on a few people and spreading the costs to many. The few have much more incentive to lobby for government programs than the many have to oppose them, so government spending grows. One-time cuts in budgets or workers won’t stop this growth.

In response to Clinton and Gore’s mandate to reduce the federal work force, government agencies eliminated workers but didn’t eliminate their positions. I remember wandering through Forest Service and Park Service offices in the mid-1990s and seeing empty desks everywhere. Each vacant desk was a memorial to some bureaucrat who lost their job, but it was also a promise to hire someone to fill the position as soon as politicians had forgotten about reinventing government. The staff who were left knew it would only be a matter of time before those desks would be filled and their pay grades restored.

According to White House data, between 1993 and 1998 the number of federal civilian jobs declined by 125,000, which was just over 10 percent, mostly due to the reinventing government initiative. But those numbers started growing again before Clinton even left office and had mostly recovered by 2003. Today, civilian agencies have about 25 percent more employees than they did when Clinton took office.

Government budgets followed the same story. The White House data show that discretionary budgets (i.e., not including mandated spending such as Social Security) declined by $30 billion between 1992 and 1996. But they immediately began rising again and by 2000, Clinton’s last year in office, they were nearly $60 billion more than Clinton’s first year in office. After adjusting for inflation, discretionary budgets today are 80 percent more than in Clinton’s first year.

Cutting spending is hard enough in the face of lobby groups that watch every line item in the federal budget. Keeping those spending cuts is even harder. For that reason, the “Department of Government Efficiency” should focus instead on making government work better by changing the incentives that lead it to spend more money.

The Federal Transit Administration’s capital improvement program, for example, rewards transit agencies for thinking up the most expensive transit projects they can, so long as those agencies can convince local politicians or voters to fund half of the project costs. So many people are innumerate that they are easily persuaded that a project that costs $350 million a mile is no more expensive than one that costs $50 million a mile, so the average cost of new light-rail projects has risen from under $50 million a mile in 1990 to more than $350 million today. Notice that there is little or no consideration for whether that spending actually improves transit ridership.

Some of that money could be cut because we really don’t need to spend more on transit. For the rest, it would be more effective to tell transit agencies that federal support must be locally matched by transit fares, not local tax dollars. If a project can increase passenger fares, which would require that it also increase ridership, then the federal government would provide funds matching those increased fares. This will force transit agencies to focus on increasing ridership and fares, not on increasing costs. (Fares should be used rather than ridership because it is easier for agencies to fabricate ridership numbers than fare data.)

I can think of similar new incentives for agencies and programs with which I am most familiar including those in the departments of Agriculture, the Interior, Housing & Urban Development, and Transportation. Such new incentives would focus agencies on the people they are supposed to serve rather than on simply increasing their budgets. More important, if properly designed these would be lasting changes, not ones that would disappear as soon as the next administration takes office. I hope Musk and Ramaswamy take note.