The utility of microsatellite markers for inferring population size and trend has not been rigorously examined, even though these markers are commonly used to monitor the demography of natural populations. We assessed the ability of a linkage disequilibrium estimator of effective population size (N_e) and a simple capture‐recapture estimator of abundance (N) to quantify the size and trend of stable or declining populations (true N = 100–10,000), using simulated Wright–Fisher populations. Neither method accurately or precisely estimated abundance at sample sizes of S = 30 individuals, regardless of true N. However, if larger samples of S = 60 or 120 individuals were collected, these methods provided useful insights into abundance and trends for populations of N = 100–500. At small population sizes (N = 100 or 250), precision of the N_e estimates was improved slightly more by a doubling of loci sampled than by a doubling of individuals sampled. In general, monitoring N_e proved a more robust means of identifying stable and declining populations than monitoring N over most of the parameter space we explored, and performance of the N_e estimator is further enhanced if the N_e/N ratio is low. However, at the largest population size (N = 10,000), N estimation outperformed N_e. Both methods generally required ≥ 5 generations to pass between sampling events to correctly identify population trend.