Single nucleotide polymorphisms (SNPs) at 12q24 are reported to be associated with breast cancer susceptibility. To date, however, attempts to identify the causal SNPs underlying the associations have been inconclusive. Here, we performed fine-scale mapping of the 12q24 locus in more that 220,000 subjects and found evidence for four independent signals, associated with estrogen receptor (ER)-positive tumor risk. The strongest causal SNPs fall within multiple cis-regulatory elements and chromosome conformation capture (3C)-based assays confirmed they physically interact with the promoters of TBX3 and MED13L. Allele-specific expression analyses identified significant associations between the causal SNPs and the allelic ratio of TBX3 and MED13L transcripts. Furthermore, reporter and allele-specific 3C assays demonstrated that cis-elements within signals 1 and 2 act as transcriptional enhancers and that the risk alleles reduced TBX3 or MED13L promoter activities. Electrophoretic mobility shift assays on representative SNPs displayed allele-specific transcription factor binding. Of these, the signal 1 risk allele disrupts a GATA3 binding site and displays allele-specific chromatin looping, suggesting this contributes to reduced TBX3 expression. Finally, we silenced the signal 1 enhancer and TBX3 by targeting a nuclease-inactive dCas9 fused to the Krüppel-associated box repressor (dCas9-KRAB) to the enhancer or promoter region, respectively. We showed that reduction in TBX3 levels by either model promoted breast cancer cell proliferation and altered cell migration and invasion. TBX3 is a member of the T-box transcription factor (TF) family, and functions primarily as a transcriptional repressor. TBX3 is frequently overexpressed in breast cancer and has been reported as a key driver of several oncogenic processes including proliferation, migration and invasion. MED13L encodes a subunit of large Mediator complex, a large multiprotein coactivator required for RNA polymerase II transcription. MED13L is reported to be required for Rb/E2F control of cell growth, the complete repression of cell cycle target genes, and cell cycle inhibition. These results establish TBX3 and MED13L as the likely target genes driving the association at 12q24, and highlight the potential importance of transcriptional repression in establishing ER-positive breast cancer.