Heteroscedasticity refers to a situation in which the variance of the error term in a regression model is not constant across observations, violating one of the key assumptions of the classical linear regression model (CLRM), namely homoscedasticity. When heteroscedasticity is present, the spread or dispersion of the residuals varies systematically with the level of an explanatory variable or with the fitted values of the dependent variable. This phenomenon is commonly observed in cross-sectional data, particularly in economic relationships involving income, consumption, firm size, or wealth, where variability tends to increase with scale. The consequences of heteroscedasticity are important from an econometric perspective. First, although the ordinary least squares (OLS) estimators remain unbiased and consistent under heteroscedasticity (provided other assumptions such as zero mean error and no perfect multicollinearity hold), they are no longer efficient. This means that the estimators do not possess minimum variance among the class of linear unbiased estimators, and therefore they are not best linear unbiased estimators (BLUE). As a result, there may exist alternative estimators with smaller variances than OLS. Second, heteroscedasticity affects the accuracy of standard errors. The usual formula for computing the variance of OLS estimators assumes constant variance of the error term. When this assumption is violated, the estimated standard errors become biased, which directly affects statistical inference. In particular, the t-statistics and F-statistics calculated using these incorrect standard errors may lead to misleading conclusions regarding hypothesis testing. Researchers may incorrectly conclude that coefficients are statistically significant when they are not, or fail to