Proponents of capitalism often seem to go against common sense. They maintain that left entirely to their own devices, profit-maximizing businesses will charge reasonable prices to consumers, maintain product quality, pay workers according to their productivity, and constantly strive to do even better. But if business only cares about the bottom line, why not charge consumers exorbitant prices for junk, and pay all workers their bare subsistence? The answer is competition. A business that gives consumers a bad deal soon has no consumers left. They take their patronage elsewhere. An employer who pays productive workers less than they are worth faces the same dilemma. A rival employer will "steal" these workers by offering them a raise. Why bother to maintain product quality? Because cutting corners is penny-wise, pound-foolish: You save on production costs, but sacrifice your firm’s reputation. Once you look at the economy through the lens of competition, the whole picture changes. To the untrained eye, greedy businesses "take advantage" of consumers and workers. Due to competition, however, the road to profit is lined with good intentions: To succeed, you must give your customers and employees a better deal than anyone else. It is easiest to see the impact of competition in markets with thousands of small firms. Economists call this "perfect competition." The wheat market is a classic example. If one farmer charged more than the prevailing price, all his customers would switch to one of the thousands of alternative suppliers. Under perfect competition, firms produce until the product price equals the marginal cost of production. Moreover, competition ensures that firms produce at the minimum average cost. Any firm with higher costs will be undercut. It is an elementary economic theorem that perfect competition’s triple equality of price, marginal cost, and average cost maximizes society’s gains to trade; it is, in economists’ jargon, "fully efficient." This conclusion is easily misinterpreted. It does not mean that only perfect competition is fully efficient. Perfect competition is a sufficient condition for efficiency, not a necessary one. It is important to bear this in mind because in most industries, perfect competition will not naturally arise, and would be disastrous to impose. Perfect competition normally exists only if the minimum efficient scale - the smallest quantity a firm can produce at the minimum average cost - is small relative to industry demand. (Figure 1) If minimum efficient scale is large relative to industry demand, in contrast, only a few firms can survive; a small firm would have enormous average costs and be undercut by larger rivals. (Figure 2) Fortunately, is entirely possible for market performance to be as good with few firms as with many. Indeed, two genuine competitors may be enough. Imagine that two firms with identical and constant marginal costs supply the same product. The firm with the lower price wins the whole market; if they offer the same price, each gets half. Competition still induces each firm to price at marginal cost. Why? If the firms initially ask for $1.00 above marginal cost, each firm could steal all of its competitor’s business by cutting its price by one penny. Like an auction, price-slashing continues until price and marginal cost are equal. What if the minimum efficient scale is so large that only one firm can survive? Surely competition breaks down? Not necessarily. As long as there is potential competition, a single firm may act like a perfect competitor. As long as other equally able firms would enter the market if it became profitable, the incumbent firm cannot raise prices above the competitive level. Note, though, that in this case competition drives price down to average cost, but not marginal cost. (Figure 3) Given fixed costs, a firm that set price equal to marginal cost would lose money. When does competition not work? The simplest and most empirically relevant answer is: when government makes competition illegal. In agriculture, governments have long strived to hold prices above the free-market level. The sector would be perfectly competitive in the absence of regulation, but many deem this outcome politically intolerable. Restrictions on international trade like tariffs do the same for domestic firms. Licensing and related regulations have kept prices above free-market levels in airlines, trucking, railroads, and other industries. The intensity of government restrictions on competition varies. In the post-war era, they were especially draconian in the Third World. Under the rubric of "import substitution industrialization," many less-developed nations cut themselves off from world markets with strict tariffs and quotas. Internal policy matched, with hand-picked firms receiving strict monopoly privileges. Such policies are in retreat, but remain a heavy burden for developing countries. Public opinion and antitrust laws tend to overlook monopolies created by the government. Instead, they focus on firms’ alleged ability to hold prices above average costs even though it is perfectly legal to compete against them. There is a simple, common, and relatively harmless way to achieve this: be the best. If the lowest-cost firm can produce shoes for $10 per pair, and the second-lowest requires $12, then the former can safely charge $1.99 more than its own marginal cost. While this is not perfectly efficient, the problem is mild. Indeed, punishing industry leaders for being the best ultimately hurts consumers by reducing firms’ incentive to leap-frog over the current industry leader. There are two other commonly-cited paths to free-market monopoly: collusion and predation. The idea of collusion is that the firms in an industry stop competing with each other. This might be achieved through merger to monopoly, a formal cartel arrangement, or an informal "gentlemen’s agreement." Under United States antitrust laws, these are all either illegal or heavily regulated. When it was legal, collusion was still easier said than done. Even if the number of firms is small, it is hard to get all of them to sign a cartel agreement, and even harder to actually honor it. As the number of firms rises, the creation of viable voluntary cartels soon becomes practically impossible. Regardless of industry concentration, though, the most fundamental check on collusion is new entry. Once all of the firms currently in the industry raise prices, what happen when outsiders notice their inordinately high profits? Existing firms could invite them to join the cartel, but then the cartel has to share its monopoly profits with anyone and everyone. But if new firms are not admitted, they will undercut the cartel and ruin the arrangement. What about predation? The idea is to condition your price on the behavior of other firms. You threaten to give the product away until you are the only firm left; once consumers have no other choice, you raise prices to recoup your initial losses. But predation, even more than collusion, is easier said than done, and there are few good examples even before the existence of antitrust laws. The hitch is that the predator loses far more money than the prey. If the predator begins with a 90% market share, it loses at least $9 for every $1 than its rivals lose. It is not enough for the predator to have slightly "deeper pockets" to outlast the prey; in this example, their pockets need to be at least nine times as deep. Other factors amplify the predator’s troubles: Rivals can temporarily shut down; consumers may stock up when prices are low, making it hard to recoup the losses; successful predators may attract the attention of large-scale entrants with even deeper pockets than their own. In sum, competition is a robust mechanism for reconciling individual greed and the public welfare. The key is not the number of firms in a given industry, but whether competition is legally permissible. More firms may reduce the probability of collusion, but that is unlikely to happen anyway. "Trust-busting" and other artificial efforts to reduce concentration tend to backfire. Industries are concentrated because the minimum efficient scale is large. Nevertheless, governments can do much to strengthen competition: They can repeal the panoply of policies designed to curtail it.