In January, they emphasized their position that the debt ceiling and deficit reduction should be considered separately. “Raising the debt ceiling now is the delayed impact of budgetary choices made in 2010 and in previous years. Although we believe the U.S. government needs to enact a number of fiscally prudent reforms, in our opinion, it’s best practice for governments to enact such reforms on a holistic basis, using the broader and longer-term perspective occasioned by debate on the budget proposal as a whole, and tying any necessary increase in borrowing authority to the budget proposal that creates the need.”

In April, there were signs of some impatience. Now they said we had another two years before we need to take action to protect our credit rating. “Some compromise that achieves agreement on a comprehensive budgetary consolidation program — combined with meaningful steps toward implementation by 2013 — could lead us to maintain the rating where it is. Conversely, the lack of such an agreement by 2013, or a significant further fiscal deterioration for any reason, could lead us to lower the rating.”

And now? On July 18th, the credit-rating agency released a “research update” saying that the United States would be downgraded unless it struck a $4 trillion budget deal in the next 90 days or so. “We may lower the long-term rating on the U.S. by one or more notches into the ‘AA’ category in the next three months, if we conclude that Congress and the Administration have not achieved a credible solution to the rising U.S. government debt burden and are not likely to achieve one in the foreseeable future.”

So in under a year, Standard Poor’s has moved us from a three-to-five-year timeframe for a plan to 90 days. The question is, “why?”

This morning, I spoke with David Beers, director of Standard Poor’s sovereign-debt division. He didn’t think there was much to explain. “We live in a dynamic world,” Beers said. “The idea that we or anybody else who does analysis around U.S. public finances would have a fixed and unvarying view over time is simply naive.”

Agreed on the dynamic-world thing, of course. But what was it, precisely, that changed S&P’s view?

In Beers’ telling, it was primarily politics. The growth outlook wasn’t any better than it had been in April, but it wasn’t substantially worse. Nor had the debt burden increased with unexpected speed. It was Washington that had unsettled them. The update was clear about this. The title was “United States of America ‘AAA/A-1+’ Ratings Placed On CreditWatch Negative On Rising Risk Of Policy Stalemate” — italics mine.

“What we’re saying now,” explains Beers, “is we question whether despite all the discussions and intense negotiations, if they can’t reach this agreement, will they be able to reach it after the election?”

Put slightly differently, if Washington hadn’t tied the debt ceiling to a deficit-reduction package, and economic policy was still being made with a minimum of fuss, perhaps S&P would be less bothered now. But the bitter disagreements of the last few months have carried a cost. By showing how much trouble the two parties will have cutting a deal now, they’ve left observers like S&P wondering if we can cut one later, either.

As they put it in their most recent research update, “we believe that an inability to reach an agreement now could indicate that an agreement will not be reached for several more years. We view an inability to timely agree and credibly implement medium-term fiscal consolidation policy as inconsistent with a ‘AAA’ sovereign rating.”

So S&P is literally saying that America is not acting like a country that deserves a AAA-credit rating. Nice job, Congress.

And having upset S&P, appeasing them might not be so simple. Beers repeatedly emphasized that he wasn’t just looking for a number. He was looking for something “credible.” And credible, in his view, was something that both parties had embraced. After all, he argued, deficit-reduction plans have to be continuously implemented over a decade or more, and if there’s not “buy-in from both parties,” there’s no reason to believe that the plan will survive the inevitable changes in political control.

You might ask whether all this matters. S&P got the financial crisis almost entirely wrong — in fact, their analytical errors, alongside those of other agencies, substantially contributed to it — so why should we listen to them now?

But the question isn’t whether S&P should be listened to. It’s whether the market will listen to them. The agency estimates that downgrading America’s credit rating would lift interest rates by 25-50 basis points. They could be wrong, but I wouldn’t want to bet on it. And if they’re not wrong, well, we actually do have a pretty good idea of what it would mean for interest rates to rise by 25-50 basis points. This paper (pdf) by Third Way looked at what would happen in that scenario and concluded, among other things, that we’d lose 650,000 jobs. We really can’t afford that right now.