Hot on the heels of Harbinger’s apparent failure to secure rights to use TerreStar’s 2GHz spectrum, which could have provided an alternative route forward while its L-band GPS interference issues are resolved, comes news that LightSquared’s first lien lenders have agreed to a modification which would give Sprint the right to pay them off and take control of LightSquared’s L-band spectrum in the event of a default. This agreement is slightly different to what I had understood that Sprint was looking for back in April, and gives Sprint far less security than pre-emption of the first lien rights would have offered.
In particular, in the event of default, Sprint would apparently have to pay off the $1.5B of first lien debt and would also presumably need to keep the Spectrum Cooperation Agreement with Inmarsat in place, which involves payments of $115M per year (increasing at an annual rate of 3%) plus additional rebanding payments over the next year. Thus the effective cost to Sprint of acquiring LightSquared’s spectrum assets would be in the region of $3.5B-$4B (the NPV of the Inmarsat payments is $2B+).
That might not be too unreasonable if LightSquared had access to 40MHz of LTE-capable spectrum in the L-band (though it is somewhat higher than the current combined price of $2.86B being offered by DISH for DBSD and TerreStar), but if the FCC declines to grant LightSquared the ability to use the upper part of the L-band spectrum, in the face of the major GPS interference problems apparently found in that part of the band, it looks utterly unrealistic for what would then only be about 20MHz of LTE-capable spectrum at best.
Its also worth noting that if LightSquared decided to terminate the lease agreement with Inmarsat, after the first five years of lease payments are made, then under current arrangements it would only have access to the Phase 1 spectrum (i.e. 2x5MHz in the lower L-band and 2x5MHz in the upper L-band). If LightSquared terminated the agreement with Inmarsat without making five years of lease payments, then the default spectrum allocation would go back to even less usable spectrum, probably resembling something like the Phase 0 allocation (i.e. only 2x5MHz in the upper L-band). As a result, there is certainly no benefit to LightSquared (or Sprint) in trying to get out of the Inmarsat agreement and expecting that there will still be 20MHz of LTE-capable spectrum that doesn’t interfere with GPS.
Thus, while the amendment to LightSquared’s first lien debt agreement points the way to a potential deal with Sprint, the GPS issues might still prevent the network sharing agreement being consummated. Of course, even if the Sprint deal does go through, this agreement also increases the likelihood that Harbinger’s equity investments in LightSquared will ultimately be worthless, because if LightSquared is unable to raise additional equity funding to pay its obligations to Sprint under the network sharing agreement, Sprint would then have a (presumably substantial) claim on the LightSquared assets before LightSquared’s equity holders saw any proceeds.