Yesterday was Tax Day here in the US… I hope that this painful exercise is now behind all of us. If not, well, I hope you filed at least an extension!
Going through the numbers, it made me think about the number of decisions involved in this annual practice. Obviously there are lots of business rules needed to make the calculations that may or may not lead to a well-appreciated refund. Many tax services in the world, like IRS in the US, already use Business Rules Management Systems (BRMS) to validate returns, flag potential errors or detect fraud. Tax systems must be super flexible to accommodate the legislations that keep changing year over year: income may be taxed differently, eligibility criteria for deductions may be more or less lenient, etc. validation rules on top could work in conjunction with predictive scores to red-flag the fraudulent returns.
My mind kept wandering further…
Simulation tools come with Tax software. You can predict the impact of adding a dependent, buying a house or taking a leave of absence. This feels right but not nearly enough… What if we could run much more sophisticated what-if scenarios? Understanding the tax formula is typically beyond people’s abilities here (which explains the success of companies like Intuit). As a result, many people often make sub-optimal investment decisions as they can’t always call their CPA for advice. Simulation tools in a way that is understandable and usable by common people would be pretty cool.
But if I was in the IRS shoes…
How sophisticated are their simulation models? Granted the Administration could estimate how many American taxpayers fall into this or that bucket regarding a given criteria. Based on that you could easily figure out how many $$ would come in the Government’s pocket with a policy change and the $$ difference on the taxpayer’s budget. If they are sophisticated they could propagate the calculations further to estimate the impact on other potential eligible deductions. This would allow the Administration to better anticipate the expected revenue.
Tax calculations contribute to the bottom-line number of course but this is not the only criteria that dictates it. Other factors would definitely influence it. Take for example unemployment. Knowing the unemployment rate is soaring, you could apply it uniformly to the various brackets or apply predictions for each one to increase the accuracy of your model. Some industries, States or demographics may be more severely impacted than others. Other factors like high rate of foreclosures, delayed retirement dates or increased charity donations due to the many earthquakes that affected the world this year could be modeled into the simulation to predict the expected overall income tax revenue. By adding to the list of external factors that could influence the bottom-line result, the model precision could be improved significantly.
Granted you do need the insight of experts, in that case a gold-star team of economists but the end goal is worth it. Can you envision the power of a prediction tool that could estimate the effect of direct or indirect legislations on the income-tax-based revenue for the Government and on the taxpayer’s side too? This year, as a result of the recession, the Government’s revenue from income taxes decreased from over 20% down to 15%. Debates over whether we need to increase or decrease tax rates would become less philosophical and more objective with tangible datapoints such as this.
I am not claiming that the economists that assisted the Government over the year did not use decision modeling techniques such as these of course. My point was mostly to illustrate the practical steps that any business could make to gain objective insight. Decision Modeling is not that complicated and the return can be enormous. It is critical though that you involve the right experts that can gauge the existence and influence of each one of those factors. Your models can be refined over time.
Simulation in itself is better than nothing, but the true value is in your understanding of the components of your decisions. Do not second guess the indirect forces that impact your business. Stop shooting darts in the dark. Take charge. Be proactive, be informed.
Learn more about Decision Management and Sparkling Logic’s SMARTS™ Data-Powered Decision Manager
