Although most economists support free trade, in the 1970s a growing number of them became increasingly puzzled by the large differences between the predictions of free trade theory and real-world trade flows. Their solution to this puzzle is known as new trade theory. One mystery was that trade was growing fastest between industrial countries with similar economies and endowments of the factors of production. In many new industries, there was no clear comparative advantage for any country. Patterns of production and trade often seemed matters of chance. Trade between two countries would often consist mostly of similar goods, for example, one country would sell cars to another country from which it would import different models of cars. One explanation, associated in particular with Paul Krugman of the Massachusetts Institute of Technology, drew on Adam Smith's idea that the division of labour lowers unit costs. Economies of scale within firms are incompatible with the perfect competition assumed by traditional trade theory. A more realistic assumption is that many markets have monopolistic competition. When a monopolistically competitive market expands, it does so through a mixture of more firms (greater product variety) and bigger firms, with bigger-scale economies. Free trade expands market size beyond national borders and so allows firms to reap bigger economies of scale, to the benefit of consumers, workers and shareholders. The upside may be greater the more similar are the trading economies. This may explain why trade liberalisation is easier to achieve between similar countries. Thus, for example, the free-trade agreement between the United States and Canada produced only minor local complaints, whereas its subsequent expansion to include the very different economy of Mexico was much more controversial (see NAFTA).