Lab experiments are a major source of knowledge in the social sciences. While the natural sciences have long used laboratory experiments to advance causal knowledge, the social sciences have been slower to adopt this method. However, in recent decades, the use of lab experiments in the social sciences has increased significantly. Despite this, many social scientists still argue that lab experiments lack "realism" and "generalizability," and that they produce data that is not relevant to the real world. This paper argues that these objections are misguided and that lab experiments are a valuable tool for generating causal knowledge in the social sciences. Lab experiments provide controlled variation, which is essential for empirical scientific knowledge. In the laboratory, researchers can tightly control decision environments, allowing for the testing of precise predictions derived from game-theoretic models. For example, in the gift exchange game, researchers can test the theory that employment relationships are governed by a gift exchange, i.e., that workers reciprocate "fair" wages with high effort. This type of experiment allows researchers to observe effort and wages and to rule out confounding effects such as multiple incentives, selection, productivity differences, or repeated interactions. The paper also discusses the limitations of lab experiments and compares them to field experiments. While field experiments may be more "realistic," they often lack the tight control of variables that lab experiments provide. The paper argues that the key issue is not whether lab or field experiments are more realistic, but rather how best to isolate the causal effect of interest. The paper also addresses other common objections to lab experiments, such as the claim that lab experiments with students do not produce representative evidence. However, the paper argues that these objections are not valid, as lab experiments can be used to study a wide range of contexts and populations. The paper concludes that lab experiments are a valuable tool for generating causal knowledge in the social sciences. They provide controlled variation, allow for the testing of precise predictions, and can be used to study institutions at relatively low cost. Lab experiments can also be used to test theories and to study institutions in ways that are difficult to achieve in the field. The paper argues that the key issue is not whether lab or field experiments are more realistic, but rather how best to isolate the causal effect of interest. The paper also argues that the use of lab experiments should be increased, as they provide valuable insights into human behavior and decision-making.Lab experiments are a major source of knowledge in the social sciences. While the natural sciences have long used laboratory experiments to advance causal knowledge, the social sciences have been slower to adopt this method. However, in recent decades, the use of lab experiments in the social sciences has increased significantly. Despite this, many social scientists still argue that lab experiments lack "realism" and "generalizability," and that they produce data that is not relevant to the real world. This paper argues that these objections are misguided and that lab experiments are a valuable tool for generating causal knowledge in the social sciences. Lab experiments provide controlled variation, which is essential for empirical scientific knowledge. In the laboratory, researchers can tightly control decision environments, allowing for the testing of precise predictions derived from game-theoretic models. For example, in the gift exchange game, researchers can test the theory that employment relationships are governed by a gift exchange, i.e., that workers reciprocate "fair" wages with high effort. This type of experiment allows researchers to observe effort and wages and to rule out confounding effects such as multiple incentives, selection, productivity differences, or repeated interactions. The paper also discusses the limitations of lab experiments and compares them to field experiments. While field experiments may be more "realistic," they often lack the tight control of variables that lab experiments provide. The paper argues that the key issue is not whether lab or field experiments are more realistic, but rather how best to isolate the causal effect of interest. The paper also addresses other common objections to lab experiments, such as the claim that lab experiments with students do not produce representative evidence. However, the paper argues that these objections are not valid, as lab experiments can be used to study a wide range of contexts and populations. The paper concludes that lab experiments are a valuable tool for generating causal knowledge in the social sciences. They provide controlled variation, allow for the testing of precise predictions, and can be used to study institutions at relatively low cost. Lab experiments can also be used to test theories and to study institutions in ways that are difficult to achieve in the field. The paper argues that the key issue is not whether lab or field experiments are more realistic, but rather how best to isolate the causal effect of interest. The paper also argues that the use of lab experiments should be increased, as they provide valuable insights into human behavior and decision-making.