Unfortunately, Mankiw is not one of the former set of economists. Therefore, his entire article is a rehash of bad advice he probably got from someone from his own business department. Let's tick down the list of uncritical thinking.
The market processes information quickly. Here is the standard paean to the efficient market hypothesis. The bad logic behind this hypothesis is that if the market was priced based information instantaneously, prices would change infrequently, because usable information about earnings or dividends comes out a few times a year. In the real world, however, prices change all the time and sometimes "inexplicably". This was the thesis of Robert Schiller's 1981 paper analyzing stock price movements: they are far too volatile to possibly be explained under the EMH. However...
Price moves are often inexplicable. Here, Mankiw refutes his first assertion, citing the very same Schiller paper. Make up your mind, dude! Price movements are always explicable. If you don't believe me, read any article about yesterday's stock market trading in the business section. They will invariably report on the stocks which had big price changes that day and tell you exactly why it happened. Of course, the real reason for a big price increase is not what the reporter's financial analyst friend said on record, but that lots more buyers than sellers showed up that day (vice versa for a big decline). And one of the big reasons why that happens has to do with the price movement of the stock itself. Economists literally cannot comprehend this kind of endogenous model that allows investors to have less-than-perfect foresight, because it is the opposite of the models they use.
Holding stocks is a good bet. After regurgitating the EFH, it's not surprising that Mankiw regurgitates the "buy and hold" mantra as well. Holding stocks is never a good bet during a bear market. It is a way to lose money. The reason investment "professionals" recommend buy and hold investing is because a) they aren't any better at investing than us and this way we don't outperform them and b) they don't want to get sued when we lose money from their bad advice. I will show you in a not-too-distant post a very simple investment strategy that is guaranteed to outperform the market. The key principle of this strategy is: SELL STOCKS WHEN THEY ARE OVERVALUED AND OWN SOMETHING ELSE.
Diversification is essential. This is another piece of bad investment advice that Mankiw could only have gotten from a licensed investment professional. Diversification reduces your returns. Always. The reason is that you'll regress to the mean. Always. Now, diversification is surely useful if you don't have the time to spend finding which sectors of the market will outperform and which sectors will underperform. This way you're guaranteed to always get the average. Of course, if you follow Mankiw's previous advice, you'll already be losing when the market goes down. But at least you won't be losing more than average, I guess. Protip: Warren Buffet does not buy index funds. Neither should you, unless you don't have enough information to do otherwise.
Smart Investors think Globally. Here Mankiw is just trying to end his article. I don't think he really understands what he is writing about, but it sounds good to him. HOW, exactly, does thinking globally help investing? Most investors generally invest in their home country because a) they have the most information about their home country and feel most comfortable making investment decisions with sufficient information, or b) investing internationally has high transaction costs or is generally unavailable to investors. Mankiw seems to think this has something to do with diversification, but most global markets are highly correlated over time, so that can't be it. The purpose of diversification is to find investments that are uncorrelated. Stocks and interest rates are somewhat uncorrelated. Stocks and gold are somewhat uncorrelated. Stocks and other stocks are not uncorrelated, be they stocks in one country or many.
So maybe, in the end, it IS a good idea not to go to Greg Mankiw for investment advice. For he'll just steer you toward investment strategies that lost people about $15 trillion this past recession alone. His article is heartening to me, though. It shows the world badly needs economists that understand investing. If you keep reading this blog, I hope I can rescue at least a few of you.