We hypothesize that the trends in income and wealth inequality in the US are related. Previous literature indicates that income and wealth follow a power law distribution. We test our hypothesis by examining the Pareto coefficients of wealth and income for the United States from 1913 to 2012. Pareto-Lorenz coefficients for income are reported in the World Top Incomes Database (WTID). Using data reported in Saez and Zucman (2014) we calculate the coefficient for wealth using the methodology employed in the WTID. We first analyze the time series properties of the two sets of paremeters. First we find that we cannot reject that both series follow a unit root. Second, we cannot find that they are cointegrated. Our analysis then focuses on the changes in the coefficients. We find that changes are cointegrated. To investigate the nature of the relationship we employ a Granger causality approach. We note that we cannot reject that changes in the income parameter do not Granger cause changes in the wealth parameter while we can reject that changes in the wealth parameter do not Granger cause changes in the income parameter. Thus, changes in the distribution of income apparently do not translate into changes in the distribution of wealth; whereas changes in the distribution of wealth do affect the distribution of income. We speculate that a possible explanation is that the flow of income is dependent on the stock of wealth. However, changing income does not necessarily translate into increased wealth inequality. For example, income may be devoted to consumption rather than wealth accumulation.