The problem with any measure from the P&L statement (such as EBITDA, EBIT and NPAT) is that they rarely represent cash flow. Cash flow is important because we like to understand returns from a cash, rather than paper, perspective. However, measuring maintainable cash flow from the financial statements can be inaccurate and difficult, especially with public companies. Therefore, we’re often relegated to using P&L measures.

With this in mind, the major problem with EBITDA is that it has no provision for capital expenditure. Capex is a major cash item that doesn’t make it onto the P&L statement because capital assets are capitalised to the balance sheet. Some capital-intensive companies have huge capex, so knowing that EBITDA is $10m may be inconsequential. That company’s capital-spend alone could be $9m, leaving cash of only around $1m (all else being equal).

The advantage of EBIT (over EBITDA) is that it somewhat accounts for capex through depreciation. This depreciation figure often represents a smoothed measure of capex since it accounts for items purchased over many years.  So, in short, EBIT is a much better measure of real earnings, even if still a little inaccurate. (These other inaccuracies come from various deviations between cash and accrual accounting.)

For most companies, the disconnection between EBITDA and cash flow is too wide to be of any real use. The only exception I can imagine is using EBITDA in the analysis of businesses within the same industry where capex to revenue ratios are similar. In those cases, it is more feasible to use EBITDA, but still not ideal.