Recent studies have shown that tree-based intercropping (TBI) systems support a more diverse soil microbial community compared to conventional agricultural systems. However, it is unclear whether differences in soil microbial diversity between these two agricultural systems have a functional effect on crop growth. In this study, we used a series of greenhouse experiments to test whether crops respond differently to the total soil microbial community (Experiment 1) and to arbuscular mycorrhizal (AM) fungal communities alone (Experiment 2) from conventionally monocropped (CM) and TBI systems. The crops had a similar growth response to the total soil microbial communities from both cropping systems. However, when compared to sterilized controls, barley (Hordeum vulgare) and canola (Brassica napus) exhibited a negative growth response to the total soil microbial communities, while soybean (Glycine max) was unaffected. During the AM fungal establishment phase of the second experiment, 'nurse' plants had a strong positive growth response to AM fungal inoculation, and significantly higher biomass when inoculated with AM fungi from the CM system compared to the TBI system. Soybean was the only crop species to exhibit a significant positive growth response to AM fungal inoculation. Similar to the total soil microbial communities, AM fungi from the two cropping systems did not differ in their effect on crop growth. Overall, AM fungi from both cropping systems had a positive effect on the growth of plants that formed a functional symbiosis. However, the results from these experiments suggest that negative effects of non-AM fungal microbes are stronger than the beneficial effects of AM fungi from these cropping systems.