Overhaul

by Steven Rattner

  In our discussions, I tried to keep to my home turf of finance and business. It was hard to challenge him on economics, certainly for me but also for bona fide economists like Christy Romer, the chair of the Council of Economic Advisers. Alan Krueger, the affable chief economist at Treasury, was perhaps the Washington economist whom Larry respected most. He'd been a student of Larry's at Harvard and had received an A+.

  I tried to decide whether Larry had always been great to work for or whether the searing experience of being pushed out as president of Harvard had changed him. In part based on what I heard from a few of his former colleagues from the Clinton-era Treasury Department, I concluded that some of his rough edges must have been sanded down. Also, he had arrived in Washington this time with a chief of staff, Marne Levine, who had both Larry's respect and very sound judgment. When he returned from a meeting in the West Wing, fuming about stupid ideas that had been put forward, Marne could calm him and keep him from flying off the handle. Larry visibly worked hard to control himself. At one meeting I attended, a junior colleague in the bleachers (the couch on the other side of his office) offered an unsolicited comment. "That's one of the silliest...," Larry began, but then caught himself and said, half under his breath, "That's the old Larry. The new Larry says, 'Have you thought about it this way?'"

  Having been a star debater at MIT, he loved to argue. In many instances, of course, he used argument to express his views and try to win others over. But in other cases, he seemed to take a contrary position just to be sure all the pros and cons were hashed out. At still other times, he appeared to argue because he wasn't sure himself what he thought. For instance, I heard him take both sides—sometimes in the same conversation—on the efficient-markets hypothesis. This principle holds that because markets are always processing all available information, it's futile for an investor to try to "beat the market." In the aftermath of the 1987 stock market crash, Larry had drafted a paper against the theory; the analysis began: "There are idiots. Look around." Yet when Harry produced a set of projections for GM showing high potential returns for a prospective stockholder, Larry invoked efficient-markets theory to push back, arguing that if Harry's projections were accurate, we would already have a line of investors waiting outside our door.

  His metaphors were vivid and so was his manner of speech. Once Diana Farrell began to offer an opinion, but before she passed the midpoint of expressing her thought, Larry interrupted to say (not harshly), "I've already considered that idea and rejected it." This so amused our younger colleagues that for weeks afterward they would say to one another, as they debated one proposal or another, "I've already considered that idea and rejected it."

  Larry was an economist, however, not a businessman. Occasionally I thought he didn't have the best perspective on financial markets or business. I wasn't sure that he wanted to be told bluntly that he was wrong, especially by a subordinate. So I would gingerly try to correct his minor mistakes (Chrysler had lenders, not bondholders) and to nudge him ever so gently toward my views on business and finance. Our discussions were the high point of my Washington experience; I would leave convinced that there could be no happier future circumstance than the chance to work for him again.

  Every few days, often in late afternoon when Larry's schedule was lighter, a delegation from Team Auto would troop over to the White House to squeeze ourselves around the cramped conference table in his office.

  That was where we first presented our bill—a mid-March back-of-the-envelope estimate of how much the auto bailout was going to cost. I wanted to have a "budget" at the earliest practicable opportunity, with the notion that we would update it as our thoughts were refined by additional work. Coincidentally, the overseers of TARP had asked to see our numbers. They were getting ready to receive the results of the bank stress tests and were evidently feeling the same dread that Hank Paulson had experienced five months earlier, that TARP might not have enough money to cover all needs. Rahm began intimating that the auto bailout might have to seek new funding from Congress, probably a nonstarter under any circumstance (as the Bush administration had found) but surely impractical on our timetable. After a few anxious days, the issue receded as early signals from the stress tests came in.

  Our first, rough estimate for the "auto caper," as Ron Bloom liked to call it, was $100 billion—$75 billion on top of the $25 billion the government had already put in. Within that total, we'd earmarked $50 billion for GM and Chrysler, $40 billion for their affiliated finance companies and for "consumer support programs" to cover warranties and encourage buyers not to abandon Detroit, and $10 billion for suppliers.

  What I found offensive about these numbers was that less than three months earlier, the CEOs of GM and Chrysler had sat before Congress insisting that less than $20 billion would solve all their problems. Economist Mark Zandi's dramatic calculation of $75 billion to $125 billion, so startling at the December 2008 hearing, had proved far more prescient than those of men who should have known their business far better than any economist or government bureaucrat.

  Larry barely batted an eye—we had just begun our work, so neither he nor we were in a position to put much certainty behind any estimate. And there was no immediate need to whittle down the numbers. The existence of TARP allowed us to contemplate committing tens of billions of dollars freely, a surreal contrast to the normal process of prying money loose.

  The toughest issue on the table was what to do about Chrysler. Could it be saved? Should it be saved? The questions were debated at nearly every meeting. Finally, on March 13, the matter came to a head. Shortly after 4:30 on that chilly afternoon (though the cherry blossoms were ready to bloom), we gathered around Larry's table and took our regular seats, I at his right and Deese at his left. Filling out the table were Ron, Harry, Diana, Gene Sperling, Alan Krueger, and Austan Goolsbee. Austan, a forty-year-old economics professor from the University of Chicago, had been an adviser to Obama from the beginning of the long march to the White House. In appreciation for his steadfast support, the President had named him to one of three slots on the Council of Economic Advisers. Perhaps in part because of his special relationship with Obama, Austan never shrank from speaking his mind clearly, directly, and forcefully. He liked to go by the numbers, using the rigorous analytical approach for which the Chicago economics department was known.

  Regarding Chrysler, there was no doubt as to what Austan (and CEA chief Christy Romer) believed. He began by recapitulating the dismal state of the auto industry that we had been living with for the previous ten weeks. He went on to address the impacts on employment of a potential Chrysler liquidation. Of course the loss of Chrysler would not mean that fewer cars would be sold in America; Chrysler's lost production would be taken up by other automakers. The question was, which ones would benefit?

  Austan and his team had concluded that Chrysler's product line overlapped much more with those of GM and Ford than with those of foreign-owned companies. In their view, most would-be Chrysler buyers would likely turn to other domestically produced brands. Thus Chrysler's demise would actually boost GM's and Ford's odds of surviving the current Great Recession.

  We all knew that Chrysler's liquidation would immediately vaporize nearly 300,000 jobs—40,000 at Chrysler and the balance among its suppliers and dealers. But Austan argued passionately that once other automakers raced into the gap, the net loss would be a fraction of that, perhaps a small fraction. The Chrysler skeptics had other strong arguments for letting the company die. It was the weak number three among the U.S. automakers. Continuing to bail it out would send the wrong signal about the Obama administration's inclination to make hard decisions when confronted by corporate sob stories.

  Harry took the economic analysis and translated it into financial implications for GM. Goolsbee believed that GM would capture more than a quarter of Chrysler's customers, or about 300,000 additional cars sold per year. That would strengthen its business and reduce the amount of taxpayer capital it would need. Harry's quick
calculations showed that if Chrysler went away, GM's pretax profit could be increased by about $2.4 billion a year. That could add more than $10 billion to GM's market value, a meaningful increment for whoever ended up owning shares.

  No one challenged the accuracy of Harry's figures or those of the Council of Economic Advisers. The case for saving Chrysler was based more on political and social reality. For example, while the net job loss might ultimately be as low as the CEA suggested, on day one of Chrysler's liquidation, those 300,000 jobs would disappear. Moreover, the job loss would be directly associated with an Obama decision, while the jobs regained would be invisible.

  In addition, the government would still face huge costs in letting Chrysler fail—for instance, Michigan's state unemployment insurance fund might go broke and need bailing out. Deese offered an argument that resonated with Larry: Given the uncertainty in our economy, it was better to invest $6 billion for a meaningful chance that Chrysler would survive than to invest several billion dollars in its funeral. (The hole in this argument was that we could invest $6 billion and still have to pay for the exact same funeral.) He and other defenders also extolled the potential benefits of the Fiat alliance and of preserving "optionality" for later moves, such as an eventual merger with GM.

  Back and forth the arguments flew. For Austan Goolsbee and Alan Krueger, the CEA's analysis was compelling. Harry, meanwhile, was particularly emphatic about the positive effect Chrysler's demise would have on GM. Goolsbee and Wilson were also concerned about Chrysler's weak product line and its ability to turn the corner, even under Sergio's strong leadership, before its cash ran out. Diana, taking the perspective of a management consultant, wondered rhetorically why the government would be in the business of saving losing companies.

  On the pro-Chrysler side, Gene reminded us that liquidation would shatter entire communities. Ron emphasized the need to preserve as many jobs as possible. Brian asked why, from a policy and political perspective, we would let Chrysler go if we had reasonable confidence that it could be saved. I saw the sense in all these diverse views and felt torn.

  Larry pressed us to attach probabilities to our recommendations and countered with odds of his own. Like Bob Rubin, with whom the concept is most closely associated, Larry is an enthusiast for "probabilistic decisionmaking," a method for weighing uncertainties. He asked how likely we thought it was that federal money would keep Chrysler alive for eighteen months (essentially past the midterm elections) and for five years (potentially through another business cycle).

  At one point, he confessed that as we gave our answers, he was discounting our probabilities based on what he thought we would say. For example, knowing that Ron was in favor of saving Chrysler, Larry lowered the probability Ron assigned to the success of the alliance with Fiat. The opposite for Harry. Plainly, Larry was loving this debate.

  I was truly undecided. The economic and financial analysis for letting Chrysler go seemed compelling to me, as did Diana's logic. But I knew I was expected to form an opinion based on all of the factors, not just the numbers. Finally, Larry turned to me and said, "I know where everyone else is on this. What I'm trying to figure out is where you are." I continued to wiggle. Deese, whose judgment I respected enormously, was looking at me across the table as if to say, "Are you out of your mind? If we have a way to support Chrysler as a viable business, we can't let it go down in this economic mess."

  Larry called for a show of hands. His question was precise: "If you assume that the probability is 50 percent or greater that Chrysler would survive for five years, would you save it?"

  Diana was unhappy with the phrasing, because she thought Larry was stacking the deck—forcing those who believed Chrysler's chances were actually slim to assume a higher probability. She had suspected that he wanted to save Chrysler and now was sure of it. While she recognized that under Larry's formulation she should be voting to save Chrysler, she voted against it anyway, as a kind of protest. Austan felt sandbagged too—he thought that if Christy had been present to vote, and if Deese, a junior staffer, had been barred from voting, a further bailout for Chrysler would have stood no chance.

  As it was, the vote was 4 to 3 against a further Chrysler bailout (with Bloom, Deese, and Sperling in favor) when Larry turned to me again. I was still unwilling to commit. Frustrated, he asked, "If you were President, what would you do?" This was an effective reminder that the man I'd signed on to serve did not have the luxury of sitting on the fence. I could not duck the question put this way, so finally, reluctantly, I cast a "51-49" vote in favor of the bailout. With that, the tally was tied, so Larry's became the deciding vote. He approached the question from a seemingly simple standpoint: Which would be the bigger mistake, saving Chrysler, which would keep a very fragile competitor to GM alive? Or letting it go, which might decimate the U.S. economy?

  Larry listened closely to the "two good companies instead of three" argument that the opponents made and also worried about giving Fiat a "free option." But in the end, for him, this calculus was simpler than his courses at Harvard had been: once he was convinced that Chrysler could get through at least a couple of years, he believed the risks of letting it go represented more of a gamble than the economy should be subjected to at this critical time. As he later described it, "You could think of it as a cheap form of stimulus at a time when the economy clearly needed stimulus." We had all come a long way from the dark days of early March when, having concluded that Chrysler was not viable as a standalone company, we would have guessed that liquidation was the most likely outcome.

  When the meeting ended, as we filed out of the office, Larry pulled Krueger aside. Quietly he told him, "I hope you understand that it's different being the decisionmaker than being an economic adviser. There are other factors I have to consider." He knew how strong the case was that Alan and Austan had made, and didn't want his star student to think he'd gone all wobbly.

  Larry recognized that Chrysler was a close enough call that the President needed to be briefed, and asked us to craft a short memo. The CEA, worried that its views wouldn't be fully represented, wanted to send its own memo of dissent. But Larry vetoed that, declaring that there would be only one memo to the President, and it would fairly portray both sides of the debate. Our memo to Obama was "litigated" extensively with Larry before being dispatched the following week.

  The memo was still being worked on and Chrysler was still on Austan's mind as he prepared for a President's Daily Brief in the Oval Office on March 18. Neither he nor Alan Krueger had heard anything further about Chrysler since the vote in Larry's office five days earlier and assumed the decision had been made to save the company. They feared that they would not be told until the last minute to avoid more disputation.

  The topic for Obama that morning was an update on Wall Street, and Austan was present on behalf of the President's Economic Recovery Advisory Board (PERAB)— nongovernmental advisers, led by Paul Volcker, whose job was to give Obama a "ground-level sense" of the economy outside what he called the echo chamber of the Washington bureaucracy. Access to the President's Daily Briefs was, of course, restricted and generally didn't include Austan. In this case, Christy had told Austan, "You can go, but only to observe on matters affecting PERAB, and you're not allowed to say anything."

  The briefing was uneventful until the end, when Austan was startled to hear Summers describe to Obama our decision to move up the announcement of a $5 billion receivables guarantee program for auto suppliers to the very next day. We'd wanted to hold off until the end of March, so that the President could announce the program with his other auto industry decisions. But the supplier crisis was real and urgent, and the pressure to respond was enormous, so we'd sped up the launch even though the mechanics of implementing the program were still far from complete. "Okay," said the President, taking it in. "That sounds sensible."

  Whoa, thought Austan. "Mr. President," he interrupted, "just be aware that the second we announce we're going to save the suppliers, everybody is goin
g to assume we're saving the auto companies too. Have we really decided that? If not, we've got to figure out how we're going to message this thing."

  "Of course," Summers said curtly, and the meeting moved on. As soon as the session adjourned, he cornered Goolsbee outside in the corridor and exploded, "You do not relitigate in front of the President!"

  "I was not litigating in front of the President," Austan shot back. "He hasn't seen that program and it has nothing to do with the financial rescue." Larry felt that the freewheeling Austan was off base. The CEA had been aware of the plans to help the suppliers, and there wasn't anything about the program that mandated a rescue of any or all auto companies. The President would still have a free hand when that larger issue reached his desk.

  To make matters worse, Austan inadvertently was never informed that the carefully balanced Chrysler memo was about to be delivered to the President at almost that very moment, as Larry had promised. Austan continued to boil based on his impression that the views of the dissenters were not being reflected, a concern that would only grow in the coming days.

  Now that it was clear that Larry wanted the Chrysler rescue to proceed, I worked hard to make sure he understood how much was still left to do on the matter and how much uncertainty remained. We had issues to resolve with Fiat, had not begun to negotiate with the UAW, and of course Jimmy Lee and the creditors were still demanding 100 cents on the dollar. We had carried out so little due diligence on Chrysler as a business that to invest now would be, by private equity standards, negligent. As smart as Larry was about so many things, I wasn't sure private-sector dealmaking was his strong suit.