A Question for the Keynesians

I have been thinking about Keynesianism a bit, and I want to get some traction on a back-and-forth you sometimes see.  I should say from the outset that I am broadly sympathetic to what I take to be the Krugman viewpoint, which I won't try to summarize here.  I am just curious about one particular nuance in the debate.

So here are the basic positions:

Non-Keynesian:  "We are not as rich as we thought we were.  We need to cut back!"

Keynesian:  "Look, asset prices might have been inflated during the bubble, but the production was genuine.  We really were producing a lot more back then, which shows that we are not up against any kind of limit to how much we can produce.  Rather, we face inadequate aggregate demand."

While I would generally agree with the Keynesian in this scenario (and while I certainly don't think recessions are a good time for the government to "cut back"), I am not sure the Keynesian response, as I have rendered it, is adequate.  I'll use a little model to explain my thinking.

The world of the model has two sectors:  agriculture and manufacturing.  Each sector makes one undifferentiated good, food or machines.  However, while the quality of food is easily observable, the quality of machines is observable only with a lag.  (Alternatively, machines last a certain amount of time, and consumers cannot perfectly predict when each machine will stop working.)

The economy starts in equilibrium, with each farmer working the optimal number of hours given his preferences for food and machines.  Likewise, factory workers are working the optimal number of hours given their preferences.  However, at some point, the population comes to believe that there has been a big quality improvement in its machines.  As a result, farmers are willing to pay a higher price for the machines.  The farmers work more hours to produce more food, so as to buy more machines.  The factory workers, enjoying higher incomes, also work harder to meet increased demand.  Everyone works more hours and consumes more.  Measured GDP rises.

But then it emerges that the machines have not actually improved in quality.  It was all a mistake.  Now farmers are only willing to pay the old equilibrium price for machines.  The production of both food and machines falls back to the old equilibrium, along with measured GDP.

At this point the Keynesian can rightly point out that the economy is capable of producing more food and machines.  After all, just a short time ago, it was doing just that.  But what the economy can't do is produce more food and machines that people are willing to pay for.  In other words, if you measure production not by kilograms but by the amount of wealth created, production was actually not so high, even during the bubble.  It was an illusion.  (Although, interestingly, agricultural production really was that high - it's just that once reality sets in, incomes aren't high enough to support that level of food consumption.  And manufacturing production, too, was higher at peak than afterward, even though it was exaggerated by the quality illusion.  I guess the way to put it is that incomes were inflated, and that based on those inflated incomes production was high.  But based on "sustainable" or "true" incomes, production was excessive.)

So what is the Keynesian to say at this point?  I think one thing to point out is that my model doesn't really have a concept of unemployment (or even of money), so we don't want to extrapolate too wildly.  But the fact remains that "we aren't as rich as we thought we were" seems as though it might be a coherent way to view the world, and you can see how peak production may be difficult to restore in a sustainable way.  You can produce that many kilograms, sure, but you can't create that much (perceived) value.

Of course, maybe the right answer is that we'll know we're up against production boundaries when fiscal stimulus creates significant inflationary pressure, and since that isn't happening there is not much reason to think this dynamic is in play.  I'm fine with that answer, but I want to be sure I'm not missing a more fundamental reason that the model isn't capturing what might happen in a typical bubble.  In other words, is there a better answer to the non-Keynesian claim that "we are not as rich as we thought we were"?