There are Dynamic models to predict complex systems, from SIER models for disease spread to Bayesian networks for medical diagnostics. But are these sophisticated, data-driven approaches applied to public policy? For instance, when a government adjusts H1B visa fees, does any institution model the likely downstream effects on university master's programs, especially those at schools that rely on international students? It seems like we're constantly reacting to policy outcomes, but are we building the computational frameworks to anticipate them? Are there governmental/think tanks that build such models, to what extend do they build it.

More general question: Compare this to a company, today we run very data driven orgs with selective feature release, A/B testing etc. Are there any tools being used at the meta level planning in Government, or even outside even in software companies or other product companies.