A 700-page treatise on economics translated from French is not exactly a light summer read—even for someone with an admittedly high geek quotient. But this past July, I felt compelled to read Thomas Piketty’s Capital in the Twenty-First Century after reading several reviews and hearing about it from friends.

I’m glad I did. I encourage you to read it too, or at least a good summary, like this one from The Economist. Piketty was nice enough to talk with me about his work on a Skype call last month. As I told him, I agree with his most important conclusions, and I hope his work will draw more smart people into the study of wealth and income inequality—because the more we understand about the causes and cures, the better. I also said I have concerns about some elements of his analysis, which I’ll share below.

I very much agree with Piketty that:

• High levels of inequality are a problem—messing up economic incentives, tilting democracies in favor of powerful interests, and undercutting the ideal that all people are created equal.
• Capitalism does not self-correct toward greater equality—that is, excess wealth concentration can have a snowball effect if left unchecked.
• Governments can play a constructive role in offsetting the snowballing tendencies if and when they choose to do so.

To be clear, when I say that high levels of inequality are a problem, I don’t want to imply that the world is getting worse. In fact, thanks to the rise of the middle class in countries like China, Mexico, Colombia, Brazil, and Thailand, the world as a whole is actually becoming more egalitarian, and that positive global trend is likely to continue.

But extreme inequality should not be ignored—or worse, celebrated as a sign that we have a high-performing economy and healthy society. Yes, some level of inequality is built in to capitalism. As Piketty argues, it is inherent to the system. The question is, what level of inequality is acceptable? And when does inequality start doing more harm than good? That’s something we should have a public discussion about, and it’s great that Piketty helped advance that discussion in such a serious way.

However, Piketty’s book has some important flaws that I hope he and other economists will address in the coming years.

For all of Piketty’s data on historical trends, he does not give a full picture of how wealth is created and how it decays. At the core of his book is a simple equation: r > g, where r stands for the average rate of return on capital and g stands for the rate of growth of the economy. The idea is that when the returns on capital outpace the returns on labor, over time the wealth gap will widen between people who have a lot of capital and those who rely on their labor. The equation is so central to Piketty’s arguments that he says it represents “the fundamental force for divergence” and “sums up the overall logic of my conclusions.”

Other economists have assembled large historical datasets and cast doubt on the value of r > g for understanding whether inequality will widen or narrow. I’m not an expert on that question. What I do know is that Piketty’s r > g doesn’t adequately differentiate among different kinds of capital with different social utility.

Imagine three types of wealthy people. One guy is putting his capital into building his business. Then there’s a woman who’s giving most of her wealth to charity. A third person is mostly consuming, spending a lot of money on things like a yacht and plane. While it’s true that the wealth of all three people is contributing to inequality, I would argue that the first two are delivering more value to society than the third. I wish Piketty had made this distinction, because it has important policy implications, which I’ll get to below.

More important, I believe Piketty’s r > g analysis doesn’t account for powerful forces that counteract the accumulation of wealth from one generation to the next. I fully agree that we don’t want to live in an aristocratic society in which already-wealthy families get richer simply by sitting on their laurels and collecting what Piketty calls “rentier income”—that is, the returns people earn when they let others use their money, land, or other property. But I don’t think America is anything close to that.

Take a look at the Forbes 400 list of the wealthiest Americans. About half the people on the list are entrepreneurs whose companies did very well (thanks to hard work as well as a lot of luck). Contrary to Piketty’s rentier hypothesis, I don’t see anyone on the list whose ancestors bought a great parcel of land in 1780 and have been accumulating family wealth by collecting rents ever since. In America, that old money is long gone—through instability, inflation, taxes, philanthropy, and spending.

You can see one wealth-decaying dynamic in the history of successful industries. In the early part of the 20th century, Henry Ford and a small number of other entrepreneurs did very well in the automobile industry. They owned a huge amount of the stock of car companies that achieved a scale advantage and massive profitability. These successful entrepreneurs were the outliers. Far more people—including many rentiers who invested their family wealth in the auto industry—saw their investments go bust in the period from 1910 to 1940, when the American auto industry shrank from 224 manufacturers down to 21. So instead of a transfer of wealth toward rentiers and other passive investors, you often get the opposite. I have seen the same phenomenon at work in technology and other fields.

Piketty is right that there are forces that can lead to snowballing wealth (including the fact that the children of wealthy people often get access to networks that can help them land internships, jobs, etc.). However, there are also forces that contribute to the decay of wealth, and Capital doesn’t give enough weight to them.

I am also disappointed that Piketty focused heavily on data on wealth and income while neglecting consumption altogether. Consumption data represent the goods and services that people buy—including food, clothing, housing, education, and health—and can add a lot of depth to our understanding of how people actually live. Particularly in rich societies, the income lens really doesn’t give you the sense of what needs to be fixed.

There are many reasons why income data, in particular, can be misleading. For example, a medical student with no income and lots of student loans would look in the official statistics like she’s in a dire situation but may well have a very high income in the future. Or a more extreme example: Some very wealthy people who are not actively working show up below the poverty line in years when they don’t sell any stock or receive other forms of income.

It’s not that we should ignore the wealth and income data. But consumption data may be even more important for understanding human welfare. At a minimum, it shows a different—and generally rosier—picture from the one that Piketty paints. Ideally, I’d like to see studies that draw from wealth, income, and consumption data together.

Even if we don’t have a perfect picture today, we certainly know enough about the challenges that we can take action.

Piketty’s favorite solution is a progressive annual tax on capital, rather than income. He argues that this kind of tax “will make it possible to avoid an endless inegalitarian spiral while preserving competition and incentives for new instances of primitive accumulation.”

I agree that taxation should shift away from taxing labor. It doesn’t make any sense that labor in the United States is taxed so heavily relative to capital. It will make even less sense in the coming years, as robots and other forms of automation come to perform more and more of the skills that human laborers do today.

But rather than move to a progressive tax on capital, as Piketty would like, I think we’d be best off with a progressive tax on consumption. Think about the three wealthy people I described earlier: One investing in companies, one in philanthropy, and one in a lavish lifestyle. There’s nothing wrong with the last guy, but I think he should pay more taxes than the others. As Piketty pointed out when we spoke, it’s hard to measure consumption (for example, should political donations count?). But then, almost every tax system—including a wealth tax—has similar challenges.

Like Piketty, I’m also a big believer in the estate tax. Letting inheritors consume or allocate capital disproportionately simply based on the lottery of birth is not a smart or fair way to allocate resources. As Warren Buffett likes to say, that’s like “choosing the 2020 Olympic team by picking the eldest sons of the gold-medal winners in the 2000 Olympics.” I believe we should maintain the estate tax and invest the proceeds in education and research—the best way to strengthen our country for the future.

Philanthropy also can be an important part of the solution set. It’s too bad that Piketty devotes so little space to it. A century and a quarter ago, Andrew Carnegie was a lonely voice encouraging his wealthy peers to give back substantial portions of their wealth. Today, a growing number of very wealthy people are pledging to do just that. Philanthropy done well not only produces direct benefits for society, it also reduces dynastic wealth. Melinda and I are strong believers that dynastic wealth is bad for both society and the children involved. We want our children to make their own way in the world. They’ll have all sorts of advantages, but it will be up to them to create their lives and careers.

The debate over wealth and inequality has generated a lot of partisan heat. I don’t have a magic solution for that. But I do know that, even with its flaws, Piketty’s work contributes at least as much light as heat. And now I’m eager to see research that brings more light to this important topic.