A key challenge in benchmarking VaR models is that trading portfolios often change over time, which violates the assumption of i.i.d. returns. This dynamic nature makes it difficult to ensure that errors from the model remain independent or identically distributed. As a result, comparing VaR model performance becomes more complex, particularly when portfolios are rebalanced frequently.

A is incorrect: While historical data availability can impact backtesting quality, it is not specific to benchmarking. Benchmarking focuses more on comparing model outputs and assumptions, rather than on data limitations.

C is incorrect: The inability to capture tail risk is a criticism of VaR models generally, but it does not directly address the challenges of benchmarking. Benchmarking involves comparing performance and robustness, not the model's inherent limitations.

D is incorrect: Consistency is a concern for validating model robustness but is not unique to benchmarking. Benchmarking focuses on relative comparisons rather than ensuring consistency in a single model.