We can't. That would require a carefully conducted cost-benefit analysis of potential outcomes including the costs and benefits of not starting it, with estimates for short-term (3 years), ten years, and twenty year outcomes. Such a study doesn't exist publicly and there is no way you can convince me it exists at all other than showing it to me with evidence that it was written before the US attacked Iran. It's also not usual to make such analyses because the costs of a human life lost are calculated very differently in each domain and are hard to assess. For instance, 13.7M per life is assumed in airline safety but that's not a figure the military would use.