Decades of research in economic sociology and network social capital established the importance of an individual’s social ties and his/her structural position in a network for acquiring valued economic resources. This strong correlation between network position and economic outcome at the individual level imply some correspondence between the distributions of network positions and income. This project broadens the scope of this theme by considering beyond an individual’s network position to the distribution of those positions and how they correspond with economic conditions at the country level. The distributional characteristics of network positions are important to consider because an abundance of particular network positions that benefit individuals who assume them may not necessarily benefit the collective as a whole. Based on a large-scale Twitter data corpus of 158M user accounts, we construct within-country communication networks and measure individual-level network social capital (e.g. network constraint, tie volume entropy) within-country. We first establish the validity of the Twitter communication network as a proxy for network social capital, despite the potential biases in the user-base and their communication ties. To this end, we estimate the correlation between an individual’s Twitter network diversity and income proxy. We develop a proxy income measure for approximately 39K U.S. Twitter users with GPS tweets, estimated from the real estate value of the most prominent location where they tweet at night. Consistent with previous findings based on more comprehensive population-level phone networks, our OLS results show a robust positive correlation between network diversity and real estate value, from which we gain confidence in the Twitter networks as capturing some signal of network social capital. On the basis of this validation at the individual level, we then conduct country-level analyses using within-country Twitter networks for 97 countries with reliable macro-economic indices. Results from random intercept models show that per capita GDP and income inequality are both positively correlated with a country’s average level of individual network diversity (i.e. higher tie volume entropy and lower network constraint). Furthermore, economically developed countries with higher per capita GDP tend to exhibit less inequality in degree distribution (i.e. lower entropy) while countries with higher income inequality tend to exhibit more inequality in degree distribution (i.e. higher entropy). This work calls for extensions in research on network social capital, from the individual unit to the population unit of analysis, aimed at theoretical developments that explain the relationship between emergent economic characteristics and the constraints on tie formation that may shape distributional characteristics of network social capital.