Instead of the traditional characterization (top 5, top field, etc.), why not use natural cutoff points? Consider the IDEAS 10-Yr recursive discounted rankings (excluding survey journals and moving AER to #5):
1 The Quarterly Journal of Economics, Oxford University Press 10.889
2 Journal of Political Economy, University of Chicago Press 6.209
3 Econometrica, Econometric Society (also covers Econometrica, Econometric Society ) 4.568
4 Review of Economic Studies, Oxford University Press 4.557
5 American Economic Review, American Economic Association 3.191
6 American Economic Journal: Macroeconomics, American Economic Association 4.136
7 American Economic Journal: Applied Economics, American Economic Association 3.521
8 Journal of Finance, American Finance Association 3.226
9 Journal of Monetary Economics, Elsevier (also covers Carnegie-Rochester Conference Series on Public Policy, Elsevier ) 3.177
10 The Review of Economics and Statistics, MIT Press 3.134
11 Journal of Labor Economics, University of Chicago Press 2.881
12 American Economic Journal: Economic Policy, American Economic Association 2.831
13 Journal of Human Resources, University of Wisconsin Press 2.717
14 Small Business Economics, Springer 2.696
15 Review of Financial Studies, Society for Financial Studies 2.636
16 Journal of Economic Growth, Springer 2.628
17 Journal of the European Economic Association, European Economic Association (also covers Journal of the European Economic Association, MIT Press ) 2.567
18 Review of Economic Dynamics, Elsevier for the Society for Economic Dynamics 2.234
19 Journal of International Economics, Elsevier 2.206
20 Economic Journal, Royal Economic Society (also covers Economic Journal, Royal Economic Society ) 2.131
Apart from AEJ: Macro (Factor > 4), there are natural cutoffs for the top 5, top 10, and top 20