Since 1993, oyster reef replenishment efforts in the Virginia portion of the Chesapeake Bay have relied heavily on construction of oyster shell reefs with enhanced vertical relief. We evaluated the performance of six reefs constructed in proximity to natural subtidal oyster bars by comparing recruit densities (spat m^-2, where spat are young-of-the-year oysters with shell heights less than 50 mm) between habitats. Recruitment was higher on the reefs than bars during the first 1–3 yr post-construction, usually by at least an order of magnitude. Within 7 yr, recruitment was similar between reef-bar pairs although both reefs and bars received additions of shell, live oysters, or both during the study period. At decadal time scales, constructed oyster reefs did not show enhanced recruitment relative to adjacent natural oyster bars. The rapid decline in reef recruitment post-construction is likely related to three processes: (i) shell degradation by taphonomic processes, (ii) biofouling that occludes the shell surface to recruitment, and (iii) inability of extant oysters on the reef to produce new shell at a rate commensurate with losses to (i) and (ii). There appears to be a requirement for continued replenishment activity to maintain the shell base on these reefs, contrary to the dynamics of a healthy natural oyster population. The similarity in recruitment between constructed reefs and natural bars at decadal time scales suggests that subtidal shell plants or shell additions to natural bars may be a more cost-effective repletion strategy because they provide equal population enhancement per unit area.