Despite successive codes of best practice of France, Germany and the UK highlighting the importance of the independence of non-executive directors, the codes tend to ignore the links that directors of family firms might have with the controlling shareholders. This is of particular concern for firms with concentrated family control as the risk of minority shareholder expropriation is greater for such firms. This paper proposes a new measure of board independence for family firms. Using a sample of listed French, German and UK family firms with an incumbent family CEO due for re-appointment or replacement over 2001-2010, we show that our measure of board independence is significantly lower than reported board independence. In contrast to reported board independence, our measure is a good predictor of the type of new CEO succeeding the incumbent CEO. Our results suggest that conventionally defined, or reported, board independence is biased and fails to provide investors, including minority shareholders, with an accurate measure of board independence. This conclusion has important policy implications for regulators and best practice in corporate governance.