Stocks, after stumbling last week, were up on Monday, in part because of comments from Ben Bernanke that suggests the Fed might be willing to do more to goose the economy. That’s sure to lift the spirits of the crowd that believe the three-year bull run in stocks is far from over. Burton Malkiel recently became the latest high-profile market commentator to proclaim that stocks are cheap. Warren Buffett and strategists at Goldman Sachs have also been out there saying it is still time to buy stocks.

Malkiel’s argument for stocks is awkward and a little bizarre for a number of reasons, not the least of which is the fact that Malkiel is the one making the argument. Malkiel, afterall, is a Princeton professor who is famous for his book, A Random Walk on Wall Street. The central thesis of that book is that you can’t do what Malkiel is now trying to do – predict where stocks will go in the future. It’s a “random walk.” But even putting that aside, Malkiel’s logic has some big flaws.

Malkiel says stocks are a good buy right now because the average dividend yield of the Standard & Poor’s 500 is about 2%. Combine that with a projection from analysts that corporate earnings will rise 5% a year on average and Malkiel says investors buying into the stock market right now can expect a 7% return on their investment for the foreseeable future. With 10-year Treasury bonds now yielding 2.24%, that sounds pretty good.

The problem is that while Malkiel is saying stocks are a “good buy” right now, his argument really has nothing to do with how stocks are priced, which would actually determine whether stocks are a indeed a good investment or not. Earnings could rise 10% a year for the next decade, but that wouldn’t help investors buying in today if stocks are already overpriced. The price/earnings multiple is what determines whether stocks are a “good buy” or not. And by saying nothing about p/es, Malkiel is essentially saying that he thinks stocks are fairly valued right now, meaning p/e multiples aren’t likely to go up or down.

Recently, though, the average trailing p/e multiple – calculated by dividing the price by the 12 months of earnings – of the stocks in the S&P 500 has been rising, after falling for much of the past decade, up from around 10 at the beginning of the bull market to about 14 now. But that may be about to reverse. That’s because all the things that typically make p/es go higher look tapped out. Interest rates and inflation, things that typically make stock values go up when they fall, are as low as they can go. And the opposite is true of profit margins, which are at an all-time high. What’s more, high U.S. budget deficits likely means corporate taxes will rise as well, a negative for stock market multiples as well. “It’s hard to expect any of the basic big things that make multiples go up to improve from here,” says Jason Trennert, a top market strategist and co-founder of Strategas Research Partners.

And while Malkiel cites a 2% dividend yield as a selling point for stocks, it’s not. Dividend yields, like p/es, are way to measure whether stocks are cheap, and 2% isn’t. The S&P 500 has average a dividend yield of 3% over the past 40 years. Buy in at anything less than that and you are maybe overpaying.

That means Malkiel’s whole argument is really rested on a prediction of analysts that corporate earnings will grow 5%. Analysts, though, are notoriously overly optimistic. They basically expect whatever happened in the past to happen in the future, especially when the recent past has been pretty good. Recently, earnings have been rising rapidly. That’s not likely to continue. And while 5% might not sound like much, at a time when the GDP is only growing 2%, and profit margins are at an all-time high, 5% looks like a stretch. Rob Arnott, for one, thinks we have hit peak earnings. So not much of a reason to buy stocks at all.