At 2008 Macro-economist have done quite a good job if compared to the macro-economic mistakes done by them before WWII, at 1929. The economists definitely improved their prediction tools since then.
In the microeconomics level, if you mean by it the stock prices in the stock exchange, all the models on the long-time failed and will fail. The fundamental value evaluation of corporations with publicly traded shares is very problematic, since it is based mainly on predictions of expected future performances. These predictions in one hand take in account the past performance of the company, but also many other more subjective factors, like management team, market share, future of the main products of the company, etc. But above all this stands the problematic position of the top managers of corporations, with publicly exchanged shares, who usually are not a majority stake shareholders, but professional managers, whose not always represent the shareholders long term interests. Adding to these difficulties many times lack of enough expertise of the brokers to analyze companies, most of the active players in the stock exchange act as short term investors, who speculate on gaining in the short term profits out of fluctuation of the share price. In such a market the share price booms and slumps of the whole market are inevitable, and unpredictable. Since so, the economic models about the share prices, even if mathematical are more interested in human behavior, than in fundamental value of the companies.