Over the past decade, the average risk score for Medicare Advantage (MA) enrollees has risen steadily relative to that for fee-for-service Medicare beneficiaries, by approximately 1.5 percent per year. The Centers for Medicare and Medicaid Services (CMS) uses patient demographic and diagnostic information to calculate a risk score for each beneficiary, and these risk scores are used to determine payment to MA plans. The increase in relative MA risk scores is largely the result of successful efforts by MA plans to identify additional diagnoses, also known as coding intensity, and not of changes in enrollees’ true health. In this article I estimate the effects of coding intensity on Medicare spending over the next decade. Under the moderately conservative assumption that coding intensity will decelerate, Medicare expenditures are expected to increase by approximately $200 billion. CMS has implemented a variety of strategies since 2010 that lessened the impact of coding intensity on Medicare spending; it has a variety of policy responses at its disposal to mitigate the impact going forward. The problem could be largely solved if CMS adjusted for coding intensity using the principle that MA beneficiaries are no healthier and no sicker than demographically similar fee-for-service Medicare beneficiaries, returning to the budget-neutrality approach that was introduced in 2004 and later abandoned.