Citi finally sells Phibro

By Felix Salmon
October 9, 2009
April, so the first reaction to today's news is "what took you so long".

Interestingly, we now, finally, get to see some numbers on just how profitable Phibro has been:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

It’s been obvious that Citi had to sell Phibro since April, so the first reaction to today’s news is “what took you so long”.

Interestingly, we now, finally, get to see some numbers on just how profitable Phibro has been:

From 1997 until the second quarter of 2009, Phibro averaged approximately $200 million per year in pre-tax earnings, while over the last five years Phibro’s earnings averaged $371 million per year. Phibro has been profitable each fiscal year since 1997, attaining profitability in 80 percent of all quarters.

If Andrew Hall was really in line for a $100 million bonus this year, that’s an enormous chunk of Phibro’s medium-term profitability. Paying Hall on the basis of average annual earnings over the past five years makes a certain amount of sense, but paying him 27% of average annual earnings over the past five years is more than a little excessive.

In any case, you can see why Citi says, in its own news release , that the numbers here “are not material to Citigroup’s earnings”: by the time Citi offsets Phibro’s annual profits (after payouts to Hall) with the amount that Oxy is paying for the company, the most important result here is that Citi has managed to lose its highest-paid employee (who owns a castle). That’s going to be very helpful when it comes to pay negotiations with Kenneth Feinberg.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

Your conclusion may be unchanged, but your metric is a questionable. I don\’t think it makes sense to evaluate a person\’s salary based on the segment\’s (after-salary) income. Much better, I think, to say that there was $471 million available to Citi and Hall, combined, after all other Phibro expenses were accounted for. Of that, 21% went to Hall, while 79% went to Citi. Is that a good split? I have no idea. But I think it\’s a better way to frame the question.

Posted by Andy | Report as abusive

Felix–I’ve yet to see anyone comment on the obvious: Citi’s promise to pay one employee $100M was under attack. Citi now believes that it has avoided the attack by pawning off the employee’s division to Phibro which, presumably, pay the employee the bonus. But, of course, we have to assume that the economic burden of the $100M bonus falls on Citi via a reduction in the sale price. Thus, Citi has avoided oversight on this issue by Feinberg.

Have I missed something?