Suppose we wish to know the effect of a change in corporate tax rate on capital expenditures. A country implements an increase in the tax rate from 15% to 20% on all corporations that had over $1 billion dollars in revenue in 2020, with the tax first being in effect in 2020. Naturally there are many things that affect a firm's capital expenditures and their revenue, and these may differ over time. Suppose we have a simplified scenario where we know exactly what firms capital expenditures will be based on the following process capex =f (revenue, year, taxrate, quality, industry) where quality is unobserved and does not change over time, and f is some arbitrary (possibly nonlinear) function. We simulate this data using the following code: We can run an OLS regression of capital expenditures on tax rate with the following line of code. We know from our generating process above that the true effect should be a decrease of $100000 per every 1% change 1 (\$500000 for the total policy change). When running this yourself set eval=FALSE to eval=TRUE if using the rmarkdown file. lm(data=df,capextax) Does the OLS regression give an unbiased estimate for the effect of a tax increase? What factors are important in answering this? What is the estimated effect of increasing the corporate tax rate by 5% on a firm's capital expenditures using bivariate OLS