EBITDA is about the closest measure you can get to in securing a meaningful comparison between companies.
The impact of "I" (ie: interest) will be felt highest in debt carrying companies such as BPC, TMS, HTA, UEC, GMF, etc, and lowest in "pure" equity plays (ie: SLX where net debt is $nil, or in SIE where there is a substantial net cash balance).
A comparative "I", therefore, is quite difficult unless you can derive an abstract WACC figure for a company, or for a particular sector. But even here, WACC will factor for the COQ (cost of equity) and COD (cost of debt). So, even by taking WACC, you will not be able to get back to NPAT.
The impact of "T" (ie: tax) will be felt highest in those companies which are essentially Australian domiciled, and producing a manufactured product.
Companies in the manufacturing and services' sectors, therefore, come into play here as the highest average taxc-burdened companies, whilst those companies spanning technology, mining,exploration, financial services, and /or international operations, will tend to have a lower overall tax burden.
For a quick comparison, perhaps you should compare the effective tax rates of NCP (low to lower), to those of FXJ (mid to high), any of the wine /retail companies, TLS (mid range), or even to some of the Banks (such as NAB) (also mid range).
In contrast, meaningful "T" comparisons are next to useless when comparing companies with substantial accumulated tax losses in place.
It is much the same argument for both the "D" (ie: depreciation), and "A" (amortisation) factors, where a higher D will be experienced in capital intensive companies which are asset rich (ie: UEC, HTA, telco companies, etc), and a higher A will be experienced in acquisitive companies (ie: those which have acquired other companies).
I, therefore, have to agree with Eddie that extrapolating from EBITDA to NPAT is not readily possible given all the vagaries involved in the make-up of each individual company.
