AbstractThis paper examines whether there is a link between establishment closure and union presence using data from the 1984–90 panel sample of establishments of the Workplace Industrial Relations Survey. It considers whether the notion that unions damage long term performance and ultimately force uncompetitive establishments to shut down is a reasonable one.I can find no evidence whatsoever for this hypothesis. There is no evidence that establishment closure and union presence are positively correlated. The relationship is usually negative and sometimes significantly so.These results are important for at least two reasons: (i) it does not seem that unions have detrimental effects that force establishments to close down in the long run; (ii) the idea that union establishments have been dying at a faster rate forms no part of the explanation of the sharp union decline observed in the 1980