Even this methodology doesn't add up. About one third of American households are in the 50 largest metro areas. If income-pegged rents increased in that group

Even this methodology doesn't add up. About one third of American households are in the 50 largest metro areas. If income-pegged rents increased in that group by 175%, but only 17% nationally, then that would imply that rents outside the largest metros decreased by 60%. That seems highly implausible. Either this study has to be wrong (or highly misrepresented) or the BLS Consumer Expenditure survey is wrong. Consider that the BLS survey is a huge sample size, with carefully controlled data, developed over decades, I'm trusting it.