We propose the notion of "peripheral penalty" to designate a negative effect suffered by a household due to a residential location at a distance from a regional metropolis. This penalty associated with a place of residence on the periphery, at a distance from the urban center, is the symmetrical of the notion of agglomeration economy, used to qualify corporate earnings, and whose role is central to the new geographic economy. The issue of peripheral penalties is part of spatial economics. It can be extended to the notion of ultra-peripheral penalty, which can be applied to ultra-marine territories, and which is more akin to regional economics. Without ignoring other areas, we focus on penalties in access to employment. To measure them, we compare the contributions of observational and experimental data and draw on the body of research that has proposed a precise measurement of these penalties, particularly on French data. We show that peripheral penalties combine both a distance effect and a location effect, and that they can be increased when the applicant is of ultraperipheral or foreign origin. All in all, households located on the outskirts of regional metropolises suffer severe penalties in access to employment, largely underestimated by spatialized literature which focuses essentially on transport costs. It seems to us that the crises of the 2000s have neither amplified nor attenuated these penalties, but that public policies have an eminent role to play in offsetting them.