This working paper studies how monetary policy transmits when a large share of households consume close to current income. We study this possibility by combining nationally representative Moroccan household expenditure surveys with a tractable two-agent New Keynesian model. As the surveys do not report liquid assets, borrowing limits, or saving flows, we use food-expenditure shares as a proxy for short-run budget tightness. At a common 40 percent cutoff, 68.8 percent of households in 2007 and 57.2 percent in 2014 are classified as budget-tight. Under the benchmark household-share mapping, these estimates are compared with a model-implied critical constrained-household share of 54.5 percent. The 2007 calibration lies inside the inverted region. The 2014 calibration sits near the boundary and returns to the standard region under stricter cutoffs, higher markups, or consumption weighting. Neither classification is unconditional. In the inverted region, passive forward-looking rules can restore determinacy, active rules require a sufficiently strong output response, and a commitment targeting rule remains determinate throughout. A stylized bidirectional exposure of constrained households to cyclical profit deviations can restore the standard demand slope. In emerging economies where budget tightness is widespread, household heterogeneity is central to the transmission of monetary policy, and the sign of that transmission should be tested rather than assumed. This record contains the main working paper and its online appendix.
ELHAMZAOUI et al. (Fri,) studied this question.