OBJECTIVES: This study examines the impact of reference pricing on the price and spending for in vitro laboratory tests for a large private employer in the United States. METHODS: We obtained comprehensive laboratory and medical claims for employees of a major US firm from 2009 to 2013.  Each claim contained the CPT code for the test, the price paid (allowed charge), and the identity and the location of the laboratory.  We matched lab claims to medical claims for the patients, obtaining information on age, gender, diagnoses and co-morbidities (ICD9). Beginning in January 2011, the firm limited its payment for 566 in vitro laboratory tests to the 60^th percentile in the price distribution for each test across national laboratories, hospital-based laboratories, and freestanding local laboratories.  Reference-priced tests generated 99,225 individual patient assays and insurance claims. Another 247 types of in vitro tests were exempted from reference pricing; these generated 33,480 individual patient assays and insurance claims.  We used difference-in-difference multivariable regression analysis (GLM with log link and gamma distribution) to measure the rate of change in prices paid per test for laboratory tests subject to reference pricing, compared to the change in prices paid for laboratory tests exempted from reference pricing.  RESULTS: Reference pricing led to a 17.8% (p<.001) reduction in laboratory test prices paid in the first year after implementation.  By the second year, this had increased to 21.6% (p<.001) and by the third year to a 27.7% (p<.001) reduction.  In the three years after implementation of reference pricing, the employer saved 18.5% in spending compared to what it would have spent had the prices of reference-priced tests continued to increase at the same rate as non-reference priced tests. CONCLUSIONS: Reference pricing reduces prices paid per test and total spending on laboratory tests in the United States.