Income inequality is undoubtedly a major challenge. In addition to the clear social and human dignity issues, excessive inequality can stifle economic growth and decrease prosperity.

While some factors influencing earnings differences are relatively straightforward (such as education), others are more complex. The degree of inequality in the United States is greater than in virtually all other developed nations. Even so, it is crucial to tread lightly with policy attempts to change the situation, lest the problem be compounded. Here are a few points to consider.

The Organisation for Economic Co-operation and Development (OECD) is currently sponsoring a forum addressing income inequality around the world. The OECD notes that income inequality has reached record highs in most OECD countries and has increased over time. Whereas in the 1980s, the richest ten percent of the population had seven times the income of the poorest ten percent, that ratio has now reached almost ten times. The ratio in the United States is almost 19. While that level is far lower than some nations (such as Mexico at almost 31), it’s well above other developed nations, particularly those with strong social welfare systems such as Denmark.

It is important to note that the sheer fact that the upper ten percent of earners in the United States do so well is not in and of itself a problem. Some nations with low ratios of high earners to low earners are characterized by a lack of opportunity rather than a healthy equality. Rewarding innovation with high earnings is a desirable attribute of a strong economy, and lists of U.S. billionaires include a number of young people who leveraged ideas into vast fortunes. The possibility of extreme rewards is crucial to encouraging the nation’s best and brightest to develop new ideas, resulting in products and services that enhance (or at least entertain) lives globally.

While the recession slowed progress, wages have risen notably over the past decades. The US Bureau of Labor Statistics (BLS) recently took a long-term look at earnings in the United States and how they have changed. Since 1979, median weekly earnings for full-time wage and salary workers rose from $733 to $791 on an inflation-adjusted basis. However, weekly earnings for the lowest 25 percent have been basically unchanged (again, on an inflation-adjusted basis) over the past 35 years. (A crucial point in any policy debate is, of course, whether people tend to stay in a particular earnings category or have the ability to move up over time.)

Another aspect of inequality, the gender pay gap, has been a topic of research and debate for decades (and the subject of a column I wrote a few months ago in the wake of an Oscar acceptance speech). The Equal Pay Act of 1963 made it illegal to discriminate against women in the payment of wages, but differences persist. Over the past few decades, women have made great strides towards pay equity which include gains in labor force participation, educational attainment, real earnings, and employment growth in higher-paying occupations.

The raw wage gap between men and women has been shrinking. According to the BLS, women’s real median weekly earnings increased 30 percent over the 35 years between 1979 and 2014. Men’s earnings over the period were virtually unchanged. For women at the upper end of the earnings spectrum (highest paid ten percent), the growth was even more impressive: 67 percent compared to a 34 percent gain for men.

Averages can be misleading in many ways, and individual choices play an important role in gender earnings. For example, if men in a particular field tend to have more experience, training, or education, it is to be expected that they earn more. Data regarding the number of hours worked show that men tend to work more hours, another factor in their higher average pay. News organizations and headline statistics tend to omit other factors contributing to the gap, including length of career, educational attainment, and job sectors. Dealing with improper discrimination is, of course, a desirable course of action, but no policy mandate will (or should) affect personal preferences and choices regarding life decisions that affect earnings.

Occupational choice is at the heart of future earnings, and education is a key driver of career possibilities. Encouraging young people to stay in school and pursue higher education and training is clearly a part of the solution, as is ensuring public education quality. Eliminating discrimination is also a worthy goal, as is facilitating upward mobility. At times, it seems that the inequality argument drifts into anti-corporate territory, but it is important that any policy prescription for improving income inequality does not excessively curtail the ability of companies to provide quality employment. Making college more affordable and accessible is good; discouraging innovation is not. After all, a good job is the best way to improve individual earnings.