the least tax-efficient. But, because of its much lower turnover, the benefit of tax deferral is much lower than Asset Class 2. Based on the example in 'Deferral Benefit," you might think that we've settled on tax drag as the best measure to determine the tax deferral benefit of an asset class. But, that's still not quite right. Though tax drag is a better measure of tax deferral benefit than tax rate or expected return alone, it fails to account for the effects of compounding on any tax deferral benefit.' As "Effect of Compounding: p. x, illustrates, when we examine the returns over a long period of time. the tax deferral benefit of each asset class changes. After 20 years, the tax deferral benefit of Asset Class 1 is almost 300 percent, confirming conventional wisdom. So, why is this the result? Simply put, the tax liability saved each year will grow at the asset class's rate of return. The greater the annual return, the more powerful the cumulative tax deferral benefit, even if the tax drag is lower. In fact, return can have such an outsized effect on the long-term tax deferral benefit, that some asset classes that have a relatively low tax drag can exhibit surpris- ingly strong long-term tax deferral benefit. New Evaluation Metrics Taking all of this together. we make use of metrics that can help evaluate the utility of owning any asset class in a tax-free environment: one for vehicles like PPLI, in which taxes should never be realized, the tax-exemption multiple (TEM) and its counterpart for vehicles like PPVA, when taxes might ultimately be paid. the tax- deferral multiple (TDM). The mul- tiples measure the combined effects of tax drag and compounding to understand the overall tax-exemp- tion (or tax-deferral) benefit over a given period. "Better Predictors: p. x, demon- strates that, compared to tax drag, both TEM and TDM more accu- rately predict the long-term tax deferral benefit of an asset class. In this example, we ass