Change in income baseline-endline

Hello everyone, I have a dataset with personal monthly income at two time points, baseline and endline questionnaire, same people responding each time. Income is a categorical variable: 0-1,000, 1,000-2,000 etc. I can see that between baseline and endline, some respondents' incomes have decreased, increased or stayed the same. How can I test if these changes are 'significant': i.e. my null hypothesis is that there are no differences between baseline and endline, and any that are are just due to chance. What would you do? TIA.