This paper studies the interaction of public and private debt in determining economic growth. Both debt variables are treated as endogenous and subject to regime switch with the interaction term being the threshold variable. Then we test whether this interaction variable causes non-linearity. We find strong evidence for a threshold effect. This threshold variable is also endogenous unlike the previous literature. Using data from 29 OECD countries from 1995-2014, the threshold effect of the interaction of public and private debt to growth is found to be negative and significant on economic growth when it reaches the level of 137%. We also decompose the private debt to household and corporate debt. It is found that the public and private debt interaction is likely to be through the channel of household debt and public debt. For a robustness check, we examine the threshold effects considering the effects of banking crises, output volatility, and institutional quality, different time periods and models.