an increase in quarterly dividend.
Its seven quarters since AGNC's board authorized a dividend increase for long suffering :-) shareholders. In that time the company has seen an increase in book value per share of 17%. Reasons for considering an increase in dividend:
1) Net investment income per share should increase in the next quarter to about $1.60(assuming Q1 spreads maintained) from $1.43 in Q1 and 30% higher than the $1.23 in Q3 2009 when the $1.40 dividend was first set.
2) Net investment income is adjusted by "Other Income" from hedge and security revaluation. This faired relatively badly last quarter at $0.175/share versus $0.74 in Q3 2009 and since then it had been steadily increasing to a high of $1.24 in Q4 2010. Obviously any return to performance relative to the dollars now involved will mean a blockbuster quarter.
3) The board could energize market PPS to provide the gap needed to attract new equity that is still priced be accretive to book. By my calculations a Market PPS of $31.75 - an increased dividend would be very helpful in driving share prices to that range.
I rest the case for
Management would like the highest SPO offer price that they can get, because it's money in their pocket to use for AGNC business purposes. They don't want to leave money on the table. On the other hand, management wants to make sure that ALL SPO shares are bought, so they can't price it too high (i.e., at or above current market price), else investors would go the the open market to buy shares rather than from AGNC's SPO.
It's a balancing act and judgement call that management has to make when issuing a SPO. That's why they get paid the big bucks.
Management typically includes that info on the first slide of their earnings presentation.
The current balance of the undistributed taxable income account is $55 million (or $.42 per share), an increase of $16 million over the end of Q4 2010.
Meaning management could issue a little over 39 million additional shares before ex-div without having to worry about funding the Q2 dividend for them, provided they had authorization to do so.
My BV numbers are soft "estimates". My aim was to assign BV values that corresponded more to "real time", e.g., using the BV reported for 4Q10 for the Dec. 28th P/B, as opposed to using the less accurate BV as reported at the end of 3Q10 (Yes, I'm still "mining" the prescient data). Given the preponderance of SPOs, It seems more appropriate to get as close to the actual data as possible. The results of your regression analysis are pretty compelling, but 33 still seems out of reach. I see the 31.50 neighborhood as more realistic, but there is a lot of real estate to cover between now and the X-d. Aside from the threat of an SPO, the macro environment may pose a potential drag on pps. (I just read an interesting article suggesting that Copper is a good harbinger for the S&P). We'll all find out soon enough.
Nope, you're neglecting the downside effect of new share issue (SPO), which has shown to be on the order of $1 down immediately upon the announcement. That (SPO) price drop is quickly recovered, however within a few days. AGNC history also shows that management has never priced the SPO at just $1.40 below current market price; just not enough incentive.
Great to have someone on the board who understands the bizarre nuances of trust taxation. I recently referenced the Sept. deadline, but I couldn't remember my source. I now suspect that it was one of your earlier posts. Thanks. As the ability to fund the dividend payment (presumably at its current 1.40) on a new issue is a condition precedent to issuing, do you know the amount of undistributed funds currently available to apply towards a potential SPO's dividends? Just curious. TIA
There is no doubt that the dividend looms large in the perceived and real gap between current PPS and price paid per newly issued share. If the dividend was 10 cents or non-existent, the $1 would probably not suffice.
Extending the argument to the other end of the quarter, when does the dividend participation reduce the necessary gap?
Example, consider a new issue the day before ex-div. Will new equity be attracted with much less discount because they get a $1.40 in value immediately? Of course this value will be wiped out on Ex-Div date - so I think in this case a dollar difference or at most $1.40 would suffice.
So lets say the shares are $31.40 the day before Ex-div. New shares are issued at $30 cost to buyers. Next day shares go down by $1.40 to $30, so the new buyers are net even on share value, but have banked a $1.40 per share - I think they feel like they have just robbed the mint.
This argument puts the max discount at $1.40 and is probably less.
But you can't ignore the earnings announcement that came out at the same time --- that dropped the price almost $2 that would not have occurred had the earnings not been posted when they did. In that case, the pre-announced SPO price WOULD have been ~$3 higher than the issue price of $27.85, as I claimed.
But that's just the last SPO; go back to three or four more and you'll see the same pattern of ~$3. The issue price of this last SPO, $27.85, was the highest issues price of the last several SPO's; it has been climbing with each succeeding SPO.
If you mix other events in with an SPO, it distorts the pricing gap that would have occurred without those other events. A more positive earnings announcement, for example, could have swung the price up several dollars per share and totally wiped out ANY gap caused by the SPO by itself.
Can't explain it any clearer than that.