Search This Blog

Wednesday, July 30, 2014

I smell a pro-rata

Now it's starting to get my goat.  In a recent appeal decision, the FCC denied appeals from 90% applicants who thought they should have gotten some Priority Two funding in FY 2013-2014.  The FCC denied the appeal.  I actually don't have a problem with that.  Here's what has me frowning: "In funding year 2013, there were insufficient funds available to fund any priority two requests."  Untrue.  There were insufficient funds to follow the established precedent of funding applicants at a particular discount level only if there is enough funding available to fully fund all applicants at that level.  But there was some funding available.  According to USAC's Q2 Fund Size Projections, there were at least $600 million ready for rollover on january 31, 2013.  Also, there was $250 million available in the FY 2013-2014 pot.  On January 24th, Chairman Wheeler blogged that "As part of our top to bottom review of E-Rate, the opportunity has opened to use existing funds to immediately begin to expand E-Rate funding targeted to high-speed connectivity to students in schools and libraries."  So on January, the FCC had $850 million available, plus whatever Chairman Wheeler wanted to leverage out of the fund.  The denial of P2 requests for 90% applicants wasn't approved by the FCC until February 24th, and the first denials weren't until March 5th.

Maybe the FCC couldn't have funded the entire $1.76 billion in Priority Two requests from 90% applicants.  But there is a procedure for that in 47 C.F.R. 54.507(g)(1)(iv): "If the remaining funds are not sufficient to support all of the funding requests within a particular discount level, Schools and Libraries Corporation shall divide the total amount of remaining support available by the amount of support requested within the particular discount level to produce a pro-rata factor. Schools and Libraries Corporation shall reduce the support level for each applicant within the particular discount level, by multiplying each applicant's requested amount of support by the pro-rata factor."

I have no objection to the FCC refusing to pro-rate.  I think pro-rating would be a disaster.  But it bothers me when they act like it was impossible for them to fund any applicants.  The FCC chose not to fund any P2 requests, and had to waive its own rules to do so.

At least I won't have to complain about this in the future: as I had suggested, in the 7R&O, the FCC abolished pro-rating (see 47 C.F.R. 54.507(f)(4) on page 134).

Friday, July 25, 2014

It's getting wilder and wilder

It looks like the 7R&O has spawned a new beast to add to the funding process hydra, the Wi-Fi-Rate chimera and the GSA minotaur.  I think I'll call it a satyr, since it's small and harmless.

A lot of NPRM commenters opined that applicants shouldn't have to file an application year after year for the same multi-year contract.  The FCC heard and created another application process for multi-year applications.  Oy.

I don't think I will ever use this form.  Because filling out an item in Block 5 to cover a multi-year contract is quick and easy.  It would be really easy if USAC built a "copy last year's 471" button.

What is required on the new form?  The FCC says in paragraph 194:
  1. "basic information identifying the applicant":  To do a Form 471, I just give the BEN and the rest autofills.
  2. "confirm that the funding request is a continuation of a FRN from a previous year based on a multi-year contract": You mean like the Item 15d checkbox on the Form 471?
  3. "identify and explain any changes in...the discount rate": You mean like editing the Block 4 that I can import from last year?
  4. ""identify and explain any changes in...the membership of a consortium": What?!  How can a multi-year contract survive joining or quitting a consortium?  Wouldn't a new contract be required?
  5. "identify and explain any changes in...  the services ordered": You mean like filing an Item 21 Attachment?
So let's see, the new form has Block 1, Block 4 and Block 5.  You know they aren't going to omit all those certifications in Block 6.  And now that we have measurable goals, they're not going to give up on that Block 2 information.  That's all the blocks.

For applicants who have only one FRN, and it's a multi-year contract, maybe this is a minor improvement, but for everyone else, it's a waste of time.

And as with real satyrs, innocents should beware.  Because if you have a new multi-year contract, you'll think that you can use this new multi-year contract form.  But no, you have to file a Form 471 in the first year, and then this new form in subsequent years.  Make that mistake, and the satyr will show its frisky side and you'll be....

That doesn't look like 20% per year

Who came up with this idea of phasing out voice service by lowering the discount percentage by 20 every year?  What they should have done is reduce the pre-discount amount by 20% each year.  It's better in three ways:
  1. Applicant don't have to keep track of two discount levels, one of which changes every year.
  2. It doesn't require a temporary re-write of the Form 471 to allow for different discount levels.  Applicants could just add 20% of the pre-discount cost to the ineligible cost in Item 23b on the Form 471.  OK, technically you'd also have to take the 20% off the BEAR and SPI, too, but why not just let people take a discount off the full amount and hit the cap 20% sooner?
  3. The effect on funding would be smoother.  Here's how.
There are two ways of looking at the effect that the phasing out will have on applicants:
  1. Funding lost: To applicants using the BEAR, the change will appear as a loss of funding.  How much smaller is the reimbursement check this year?
  2. Cost increase: To applicants getting discounts, the change will appear as an increase in cost.  How much bigger is my phone bill this month?
How strongly the effect is felt will depend on the applicant's discount level and invoice mode.  For example, a 90% applicant filing a BEAR will get a reimbursement check that is 22% smaller, but a 90% applicant receiving discounted bills will see it's phone bill triple (it's share will go from 10% to 30%).  Meanwhile a 20% applicant sees 100% of it's funding disappear, but the phone bill only goes up 25% (I know it seems like it should go up 20%, but do the math and you'll see it's 25%).

Here's a table showing the effect for all applicants over the next 3 years.  (Sorry, the 4th and 5th years won't fit.)
Cost Allocation Year 1 Year 2 Year 3
Original Discount Funding Lost Cost Increase Funding Lost Cost Increase Funding Lost Cost Increase Funding Lost Cost Increase
90% -20% 20% -22% 200% -29% 67% -40% 40%
80% -20% 20% -25% 100% -33% 50% -50% 33%
70% -20% 20% -29% 67% -40% 40% -67% 29%
60% -20% 20% -33% 50% -50% 33% -100% 25%
50% -20% 20% -40% 40% -67% 29% -100% 11%
40% -20% 20% -50% 33% -100% 25%
25% -20% 20% -80% 27% -100% 5%
20% -20% 20% -100% 25%

The "cost allocations" columns show the apparent funding effect if the 20% is taken off the pre-discount cost.  I didn't have to do columns for every year, because the effect is the same every year.  I didn't really have to do rows, either, since every applicant sees the same effect every year.  Looks fair, doesn't it?

The other columns show the varying effect that applicants will feel as their discount percentage is cut 20%.  Year 1 shows the strongest effect.  Does it seem fair that applicants at the lowest discount level lose 100% of their funding?  Does it seem fair that applicants at the highest discount level see a 200% increase in costs?

Why would the FCC do it?  Well, the FCC is trying to cut voice costs.  What I call "Funding Lost" is, from the FCC perspective, "funding moved from voice to broadband."  Look at that column: for every applicant in every year, it's higher than 20%.  That means that by taking the 20% off the discount level instead of taking 20% off the pre-discount cost, the FCC has accelerated the movement of funding from voice to broadband.

Too bad it whipsaws applicants.

Thursday, July 24, 2014

Meet the new rule, worse than the old rule

Oh, now I get it.  The 7th Report & Order is out, and I'm still working my way through it, and I think I finally understand what they mean when they say you have a 5-year budget for "Category 2" (née Priority 2) expenses.  I thought you'd apply in 2015, get your $150/student, then spend it any time in the next 5 years.  I was wrong.

If I'm reading the order right, if you apply for any C2 funding in FY 2015 or 2016, then you have a total of $150/student (or $2.30/sq.ft. for libraries) to spend over the next 5 years.  Whenever you need some C2 equipment, you apply, and if you're approved, the cost of that equipment is deducted from your $150/student budget, and if there's money left, you can apply for C2 again next year.  There's a good example at the end of paragraph 105.  And you're exempt from the 2-in-5 Rule, at least for the next 5 years.  Let's call the new rule "the $150-in-5 Rule" (sorry, libraries).

Rationing C2 money is just bad thinking.  If your incentive is encouraging applicants to spend too much, cut your incentive.  But you know what?  $150-in-5 could be better than 2-in-5.  Except for the bookkeeping nightmare.  And applying the cap to individual schools makes the bookkeeping much worse.  Instead of keeping track of whether a particular location has gotten funding in the last 5 years, you have to keep of how much money that location has gotten.  And I guess consortia have to keep track of every location of every member.  That's not simplifying the process.

But it gets worse.  Much worse.

First of all, what about the district's WLAN controller?  You're not going to put a controller in each building; you'll get one or two for the whole district.  Let's say you decide to buy a new WLAN controller and a new Internet router for the district.  So you plunk down $48,000 for the WLAN controller and $12,000 for the router to cover your 10 schools, the district office and a maintenance garage.  Well, it looks like you're supposed to divide the cost of the locations, so you allocate $5,000/location; the schools' cost is eligible for E-Rate funding.  If you're a 90% district, you'll get 85% of that $5,000, or $4,250/school, for a total of $42,500.  And then for each school you'll have to keep track of that number to see how much you have left on your $150/student allotment.  And if one of your schools has fewer than 29 students, or has used some of its cap elsewhere, then you wouldn't get the full discount for that school.

But wait a minute, the maintenance garage has one computer connected to the Internet, but doesn't have a wireless access point, so now we allocate the $48,000 across 11 locations since the WLAN controller doesn't serve the garage, but the $12,000 router covers 12 locations, so now the pre-discount amount allotted to each school is $48,000/11+$12,000/12=$5,363.64.  And then after 2 years, the maintenance crews get tablets [yes, they will] and they need Wi-Fi in the garage, so you install an access point.  Oh, crap, now you have to give money back because the schools got funding based on a 1/11 share, and now they're only using a 1/12 share.  If you're head's not ready to explode yet, think about the math when your statewide network upgrades equipment at the core.  Suddenly the budget calculations of every district in the state are affected by whether I connect my maintenance garage to the Internet.

Second, what about changing enrollments?  OK, paragraph 115 covers that.  The good news: if enrollment at a school drops, you don't have to give the money back.  [Got a school closing in the next couple of years?  Think of it as a Wi-Fi equipment factory.]  More good news: students that split time between districts (think vo-tech) can be double-counted.  The bad news is for new charter schools.  A new charter high school will frequently start up with just the 9th-grade class, then the next year accept a new 9th-grade class, and so on.  It takes 4 years to reach full enrollment.  Under this plan, they'd have to install their Wi-Fi infrastructure in 4 stages.  That's costly and disruptive.  And they have to make a guess about the availability of C2 funding in 4 years when they're planning the network.  What if I have a school with 500 kids and I get my $75,000 in FY 2015, and then my enrollment climbs to 600 by FY 2017.  Am I eligible for another $15,000?  Not screwy enough for you?  How about this: a district is opening a new school in FY 2017.  The district is allowed to estimate how many kids will be transferred into the new school, but those kids were already counted in in calculating the cap for the existing schools they were crowded into in FY 2015.  If it's a replacement school, the district will get $150 for each student for both the new school and the old school, and then just tear the equipment out of the closing school and distribute it around the district as needed.

Third, not every school's needs are the same.
  • How much Wi-Fi does a pre-school with 200 kids need?  Just pay your cable company $200/month and they'll throw in a wireless access point; whatever district resources they need, they can get over the Web.  Maybe you need to put a second WAP at the far end of the building; splurge and buy 802.11n, though 802.11g would be more than enough.  Total cost?  $125 for the WAP, $200 for the cable run.
  • How much Wi-Fi does a 200-student technology charter high school need?  Each kid will have a school-supplied device and a BYOD, so you'd better have at least 10 WAPs, and you're going to need some other drops in each classroom, so you're looking at a 48-port switch (and since you'll want 802.11ac WAPs, you'll need 10 Gbps ports on that switch), and those kids change classrooms every 40 minutes, so you'll need a WLAN controller.  And you'll need access to the district WAN, so you'll have to install a router.  With cabling and installation, we're over $40,000.  And we haven't yet installed a lab for these kids to work on technology yet.
  • How much will E-Rate support in those locations?  $30,000 over 5 years.
Why would they do all this?  Why?  Actually the order tells us why, starting in paragraph 108:
  1. "greater predictability"
  2. "this approach maintains the E-rate program’s priority for the highest poverty schools and libraries...  At the same time, this approach guarantees a broader distribution of funding for internal connections..."
  3. " promotes cost-effective purchasing"
To which I say:
  1. What predictability?  I am pretty confident that all 90% applicants (or should I call them 85% applicants or 90%/85% applicants, or 90%/85%/70%-50%-30%-10%-0% applicants to include their voice discount) who choose to apply in the next 2 years will receive C2 funding.  I'm not so confident for 80% (80%/80%/60%-40%-20%-0%?) applicants.  40% (40%/40%/...forget it) applicants?  Don't waste your time.  And even for 90% applicants, there is confidence of funding for the next 2 years, but after that, it's dicey.  C2 is still a lower priority than C1, and once we've spent this one-time $2 billion leverage, where will the funding come from?  Savings?  Spare me.
  2. I don't feel like I need to rebut this; the two sentences I quoted rebut each other.
  3. My favorite part of the order so far.  The text says that setting a budget for C2 funding will keep applicants from choosing more expensive C1 options, and then the footnote has comments which all say that limiting C2 funding causes applicants to choose more expensive C1 options.  Setting a 5-year budget of $150/student, but then not providing enough funding to support that, is worse than the 2-in-5 Rule: applicants will overbuy right now, then try to shove as much functionality as possible into C1 in the future.
The $150-in-5 Rule would be a little better than the 2-in-5 Rule if the Commission made 2 changes:
  1. Guarantee at least $5.5 billion for C2 over the next 5 years.
  2. Make the $150/student-$2.30/sq.ft. budget apply to organizations (billed entities), not locations.  Take total enrollment per district and total square feet per library system.  It drastically simplifies things, and removes the stupid idea that a pre-school needs as much Wi-Fi coverage as a high school.  And stops student-transfer shenanigans.  Also, with the per-district discount, there is now no reason for the complicated equipment transfer rules.
Note I said "better than the 2-in-5 Rule."  Not good, just better than one of the worst rules in the program.  As it stands now, $150-in-5 is worse than 2-in-5.  Astounding.

Wednesday, July 23, 2014

My growing bag of tricks

Hey, another Stupid USAC Web Tool Trick.  [Wait, that name makes it look like I think USAC is stupid.  I would never say that publicly.  I meant that the trick is stupid.  The tool is also stupid.]

Do you hate the Entity Search tool as much as I do?  Why can't I put in more than one Search Criteria?  Why can't I use a wildcard in the ZIP Code or Entity Number search boxes?  Why can't I choose what columns display on that first table that appears?  (At least the state appears, which is more than I can say for the Search Posted Form 470 tool results.)

So I was wondering this morning what characters would make the tool flake out, and I discovered a very cool, if not terribly useful, feature.  Check out what you can do with that % wildcard in the Entity Name field.  The instructions on the screen say "please enter at least 3 characters before the % sign."  But the real rule is: "Your query string must be at least 4 characters long, including any wildcard(s)."  Did you notice the (s)?  That's right, you can use more that one wildcard.  So, for instance, if you put in "f%dis%" (it's not case-sensitive), it will give you every entity that starts with "F" and has "DIS" somewhere later in its name.

Cool, no?

Of course, I took it to the extreme.  Yes, %%%% is a valid query string, and will return everything.  Well, it would return everything if it didn't make USAC's servers choke on the whopping list.  But by using %%%% and selecting entity types, I can tell you that there are 21,965 libraries in the database, and 17,572 districts.  And that there are enough schools to choke USAC's buffer.

Alas, this trick does not work on the SPIN Search tool, where it would really be useful in searching for DBAs.

Tuesday, July 22, 2014

Vaporware

Where is that 7R&/EMO?  The E-Rate 2.0 release date is slipping.  That's what they get for trying to add and remove features after beta (the RFC in March).  Some of those feature changes are going to cause problems:
  1. The five-year C2 budget was a fine idea, but it's going to hamstring competition if changes aren't made.  
  2. Wimping out on the reduction to the top discount level creates a good deal of complexity without saving much money, or changing the perverse incentives.  
  3. Phasing out phone service by cutting the discount level was not the simplest or fairest way to go about it (it triples costs for 90% applicants, and reduces funding for 20% applicants to zero).
  4. Sunsetting the per-student funding cap after 2 years really complicates the calculation for applicants: do I take the $150/student bird in the hand, or wait 2 years for the uncapped funding bird in the bush?
Maybe we would have been better off if the Commission had just released a bunch of patches to the program.  Because what we ended up with doesn't really deserve the title E-Rate 2.0  More like E-Rate 1.2.7.  Or maybe we should give the new release a name: Vista maybe?  Or how about Edsel?

Friday, July 18, 2014

Where have all the dollars gone?

I saw a Funds for Learning analysis applying the purported changes in the new reform proposals to past years to see what effect it would have had.  Of course, I jumped right to the graphs.  But in the first graph, I was distracted by the numbers, because they didn't add up.  FFL has total P1 demand for 2013 at $2.551 billion.  But USAC's 2013 Demand Estimate said the total P1 demand was $2.709 billion.  Why are those numbers different?

Then I decided to check FFL's numbers for broadband vs. non-broadband.  So I went to the table of funding by category that USAC sent to the FCC in June.  I dumped it into a spreadsheet to play with.  The total of the requests on that table is $2.103 billion.  Somehow that table says requests were 22% less than the Demand Estimate.

Let's see what the Data Retrieval Tool (DRT) says.  The total ($2.719 billion) is just a little higher than in the Demand Estimate.

Say, it looks like about $452 million in P1 funding was denied, and $139 billion in P1 funding is still in review, for a total of $591 million.  That would explain the difference between the Demand Estimate and the recent data estimate. But the that isn't what USAC said the table was.

Hmmm....

Generally, it's easy to say what's "broadband."  OC-1 (50 Mbps), yes.  E-mail service, no.  But some things are a bit nebulous.  ATM and Frame Relay, for instance, are variable-speed.  And what about T-1s?  I think 1.5 Mbps falls short of most standards for "broadband," but you could bond them....  Is the FCC really going to toss T-1 lines out of the program?  If not, will they toss out PRIs?  They're basically specialized T-1s.

Leaving all the questionables in the "broadband" category, I find that demand will drop about 32%.  Taking out all the questionables, demand will drop by 40%.  (FFL, by the way, estimated 35%.)

If we use that 40% number, 2013's $2.7 billion in demand would have been only $1.6 billion.  And if USAC reductions and denials were 8.5% of the total (which is where they are now), approved funding would be $1.5 billion.  So if we get a little inflation over the next few years, we could make it to the $1 billion/year that we'll need to fund broadband to provide C2 funding once every 5 years (assuming $150/student and negligible funding for libraries and the continuing non-participation of most private schools).

But spending on broadband is not going to stay flat, especially if you take away funding for T-1 lines.

The numbers only work if we're unrealistically optimistic and ignore monkey wrenches like voice not being fully phased out for years.

Popping a cap in competitive bidding

[Note from the future:  Now that the Order has been released, I see that I was mistaken about the 5-year budget.  I thought you just applied for the $150, then spent it over 5 years, but the reality is that the FCC has temporarily replaced the 2-in-5 Rule with a $150-in-5 Rule.]

I was trying to wait for the 7R&/EMO to actually come out before commenting further on policies, but I can't get this new five-year C2 budget out of my head.

It's such a good idea.  Applicants should always have gotten a chunk of cash that they can spend any time in the next 5 years.  Too bad it's a rationed chunk, but the per-student/per-sq.ft./per-toilet [tee hee! I got to say "toilet" again] cap is a separate rule.  The multi-year budget is a good idea.

But I don't see how it can work.

One positive outcome of the 5-year budget is the death of the tech plan.  The FCC couldn't have kept a straight face when saying, "We'll give you equipment funding for the next 5 years, but your spending has to be based on a tech plan, which can't be longer than 3 years, because no one should be planning tech purchases more than 3 years out."  I've been calling for the abolition of the tech plan since 2007.

But I can't imagine how the FCC is going to make the 5-year budget work with it's competitive bidding rules.  Are they just going to hand out $150/student (or $2.30/sq.ft. or $6,000/location or whatever per-location minimum they end up setting) without any list of equipment or competitive bidding?

The alternative is to force districts to guess all the equipment they'll need over the next five years, then bid it out and lock in with one service provider to come up with a fictional budget.  Then as the years go by, every time the applicant wants to actually buy something, they have to do a service substitution because the equipment they listed years ago is at best outmoded and at worst no longer for sale.  And unless the FCC changes the heinous SPIN change rules, the service provider will be able to set whatever price they want, protected from even the possibility of competition.

In a nutshell, the applicant is listing a bunch of equipment they won't end up ordering, getting pricing on that, picking a service provider based on that, and then later paying whatever that service provider wants for a different list of equipment.

Well, they could force applicants to file a Form 470 for all service substitutions, and allow SPIN changes based on that.  I suggested that as a work-around to the 6R&O SPIN change rules back in 2011.

Wait, a 470 every time you want to do a service substitution?  What am I saying?

Here's what should happen, sort of following a purchase process I suggested earlier:
  1. Applicants come up with a list of what they intend to buy, so we can pay the employees of a for-profit subsidiary of the phone companies' association (that's who runs PIA) to discover that SRST module on your router and reduce the funding request by $100, which will be totally pointless, since all my clients will be submitting requests costing at least $300/student, so PIA can chop all they want, and we'll still hit the $150/student cap.
  2. Applicants get a quote on the equipment, and based on that quote, USAC allots them a budget.  Unless the applicant is foolish, that budget will be $150/student.
  3. Don't force applicants to sign contracts.  The E-Rate process already forces applicants to sign illegal contracts, but forcing them to sign contracts for equipment that won't be delivered until years later is really too much.
  4. When it's time to buy the equipment, the applicant gets some quotes and picks a vendor.  (You want to force them to file a Form 470, fine, but that only drives up costs.)  
Now the question is, why are we going through steps 1 and 2?  In the hopes that some hapless applicant, trying to guess its equipment needs for the next five years, will come up with an amount less than $150?  Or that some applicants will foolishly come up with a list of equipment based on the $150/student budget, and then we can pay PIA to force them to cost-allocate out ineligible features so that the budget drops below $150/student?  Increase administrative costs so we can punish the honest and naive.

Just give out $150/student and $2.30/sq.ft. until the money runs out.  True, we wouldn't reach the vast majority of applicants, but that will be true anyway, until the Chairman decides an increase in the size of the fund is warranted.  Then have PIA (or the new Strike Force) bring down the hammer in step 4.

And here's an unintended benefit: more libraries will apply.  Right now, libraries don't apply because either 1) they don't want the hassle, or 2) they don't want to surrender their patrons' rights to comply with CIPA.  Since taking away is a stronger incentive than giving, if you give libraries $2.3/sq.ft., but tell them you'll take it away if they don't knuckle under to CIPA and apply, library participation will increase.

Wednesday, July 16, 2014

Still too close to free

How much did we save by cutting the top discount for C2 to 85%?  How much more would we have saved by cutting it down to 80%?  We can only estimate, because we don't know how the new district-level discounts will play out (will most 87% applicants become 90% or 80%?).  But you can use this spreadsheet to come up with your own estimate.  I based my calculations on all applicants at 87% or higher becoming 90% applicants, while 86% or lower dropped to 80%.

So based on my assumptions, dropping the top discount to 85% saved  $52,063,763.  Dropping the top to 80% would have saved twice that much.

In my mind, there is no way that saving $52 million was worth the hassle of creating a second discount level.  It will probably cost that much just to add the necessary functionality to Block 4 of the online Form 471.  Wait a minute, USAC only has 6 months to add that functionality to the Form 471, and don't they have to wait for OMB approval?  What will we do if the new form isn't ready?  And they also have to add in the 20% drop in the discount level for voice services.  Except my VoIP comes bundled with my bandwidth, so my service provider is going to have to apply two different discounts to different parts of that one charge.  And are PRIs digital transmission, since they are just a channelized T-1, or are they voice, since the only application I've ever seen is for voice; can I get funding for the T-1 if I can cost-allocate out the channelizing?  Except that because of telecom regulations, the wholesale price of a T-1 is lower if voice travels across it, so my cost allocation might result in the eligible cost being higher than the actual cost.  And what about analog data circuits?  They are data transmission, but not digital transmission.  What discount level do they get?  And if voice equipment is being phased out, do I take the 20% decrease off the new 85% discount for C2?  Now I have 4 discount levels for each 90% applicant.

Wait, where was I?

Oh yeah.  $52 million saved at the expense of a considerable complexity added to the application process.  Not a good buy.

On the other hand, this could be the camel's nose that will eventually lead to an across-the-board drop in the top discount level to 65%, like over at the Rural Health Care program.  If the top discount were 65%, we'd save  $682 million, just for C2 (using the $150/student cap).  But more important, we wouldn't need all the stupid rules that are only necessary because 90% is too close to free.  And so is 85%.


Cap ↦ Law?

Commissioner Pai said that Chairman Wheeler has been promising "outside parties" that he'll increase the E-Rate fund.  Representatives Upton and Walden said they were troubled by those allegations.

Now it seems there is a possibility that an amendment is coming to the House floor that will limit the FCC's ability to increase the size of the E-Rate fund.  Man, I hate it when politics come to the E-Rate.

I'm no political expert, but faithful readers know that a lack of expertise never stops me from shooting my mouth off.  So here's my thinking.

First, I don't see this amendment getting through the Senate.  So nothing is going to happen until after the election.

After the elections, look for an FCC decision on raising the cap.  Right now, an increase seems likely, but I'm not betting my life savings on it.  How much of an increase?  Well, Commissioner Rosenworcel has said: "At a minimum, we need to...[bring] back what inflation has taken away....  But we should go beyond this...."  According the Bureau of Labor Statistics, if the E-Rate had been indexed since 1998, it would be $3.28 billion now.  And hey, that's about the same as the $1 billion/year that the Chairman seems to think is enough to get where he wants to go.

After January, my analysis splits.
  • If a Republican wins the White House, I'll bet Commissioner Pai's ideas on the E-Rate get more attention.  Within a year or two, we'll be looking at a tightly capped E-Rate and a per-student allocation for everything.  Eventually the Tea Party will notice that a fee on their phone bill is being used to give schools a per-student subsidy, and then who knows.
  • If a Democrat wins the White House, the increase will probably stick; even if the Republicans get a majority in the Senate and a "cap the E-Rate" bill passes, there won't be enough votes to override a presidential veto.
I'd feel a whole lot better if the increase had been decided now (or 6 months ago).

Monday, July 14, 2014

I'm not shouting GOOOOOOOAL



Most of the reactions I've read are full of platitudes like "good first step" and "squandered opportunity."  Among politicians, the choice of platitudes seems to fall along party lines.  The reaction from applicants has generally been in the "good first step" camp.

I think the 7R&O/EMO does more harm than good.  Let's look at whether it meets the goals the Commission put forward in the statement released Friday:

Closes the Wi-Fi Gap: The 7R&O/EMO takes 3 concrete steps to increase Wi-Fi adoption: 1) leverages the fund to free up an extra $2 billion, 2) tosses other services out of the program, and 3) caps C2 spending at $150/student ($2.3/sq.ft. for libraries).  There is widespread consensus among program participants and the American public that this program should get more money, but leveraging the fund is not the way to do it, especially since it only gets us 2 years of funding.  I suggested tossing voice out of the program over a year ago, so I have no problem with trimming the Eligible Services List.  The C2 funding cap is just another in a string of bad rules to try to overcome the perverse incentives created because 90% is too close to free (and so is 85%).
In short: Focus on Wi-Fi and broadband, good.  Creating more perverse rules to compensate for perverse incentives, bad.  Only enough funding to get Wi-Fi to 30% of students, even if we stiff libraries.

Maximizes E-Rate Spending: There is no evidence that bulk buying lowers prices, and creating incentives for bulk purchasing will encourage applicants to join bulk buying even when buying individually would be cheaper.  Transparency is very good.  Did someone say Lowest Corresponding Price?
In short: Increased price transparency should more than overcome the pricing inefficiencies of bulk buying.

Simpler Application Process: The streamlined process for multi-year applications is apparently only for some multi-year applications, and really, if you're doing a 471 already, the effort of adding a normal Block 5 item for your multi-year contract is negligible.  Apparently at first, the small-dollar, cost-effective FRNs with an expedited process will be only "business class" Internet access (around $200/month for 100 Mbps best-effort). So now if I have a multi-year Wi-Fi contract, low-dollar Internet access and month-to-month phone service, I have to keep track of 3 different application processes?  And 2 different discount amounts (3 if I'm a 90% applicant).  How is this simpler?  And I have to file electronically, regardless of whether I find that simpler or more difficult?  And now I have to amend my district's document-retention policy to include keeping E-Rate-related documents for at least 10 years, hoping that won't conflict with state rules?
In short: Faster application processing is good, but it's not due to this order.  The other reforms just make the application process more bewildering and onerous.

Excellent: pricing transparency 
Good: focusing on broadband
Bad: everything else
Good in a bad sort of way: This order will drive more applicants into the arms of E-Rate consultants, so my personal business outlook is good.  It's confusing enough just trying to guess the name and acronym for the 7R&O/EMO.

Please, don't strike our schools

Really?!  You're sticking with "Strike Force"?  I complained about that bit of rhetoric when I first heard it, but now it's official: the FCC actually created a division called the USF Strike Force.  For now, I'll treat it as just posturing, but if I hear about the FCC buying drones....

Laws v Rules

In his statement on the E-Rate Modernization Order, Commissioner O'Rielly said, "I do not see where lowest corresponding price is reflected in the underlying statutory authority."  Well, that got me digging.  Sure enough, the three words "lowest corresponding price" are not among the 492 words of 47 U.S.C. § 254(h)(1)(B).  (There were 492 words in that subsection of the law in 1996.  Thanks to CIPA, that subsection is now over 3,000 words.)

The statute also has no hint of anything about competitive bidding, or reimbursements to schools when carriers don't discount bills, or funding for equipment, or limiting funding for equipment to twice in 5 years, or funding services but not equipment to administrative buildings, or record-keeping requirements, or anything in PIA's 700 pages of secret rules, or record-keeping requirements, or just about any other rule in the program. I don't see that the E-Rate program as it exists at all resembles the sketchy outline in the statutes.

The FCC has made up lots of rules that are not reflected in the statute. Just Subpart F of 47 C.F.R. 54 is over 12,000 words, and that doesn't count other applicable sections of Part 54, to say nothing of: dozens of appeal decisions, some with rules tucked into footnotes, the 700 pages of secret rulesold USAC Web pages and slides from old USAC presentations that the FCC has treated as rules.

So why single out lowest corresponding price?  I'm all for reducing the number of rules in this program, but Lowest Corresponding Price is not where I'd start.

Saturday, July 12, 2014

Winners and losers

Who wins under the E-Rate Modernization Order?  Who loses?  It's too early to say conclusively, but based on what we know, I can make a few solid guesses.

Winners:
  1. 90%-discount applicants.  The Priority 2 gravy train had come to an end.  Now the FCC has sacrificed everyone's telephone reimbursement and leveraged the fund (meaning it no longer has enough money to pay its obligations) to create enough funding to give 90%-discount applicants one more chunk of P2 (sorry C2) funding.  80% applicants may get in on the party, but that's not a sure thing.
  2. Data network equipment vendors.  The P2 spigot had run dry, and now the FCC has turned it back on for a couple of years.  And they don't have to share with those pesky PBX vendors.  OK, so the data vendors also sell voice equipment, but still, they've got to be happy that voice is gone.  And the 2-year limit on funding means that applicants who bought 802.11n access points last year are going to buy 802.11ac access points next year, because who knows when they'll have another chance.
  3. Big phone companies.  You thought they'd be on the list of losers, since voice got kicked out of the program?  I think the phone companies are winners for the following reasons:
    1. The E-Rate will drive applicants away from traditional voice, which at this point is just a pain in the neck for the baby bells.  If VoIP is out (and it seems it is), that will sting a little, but not really.
    2. All those pesky services like webhosting and email that were offered by non-carriers get driven out of the program, leaving only services offered by telcos in Category 1.  
    3. With voice hardware out of the program, telcos will get more of that lucrative hosted phone service.
    4. The emphasis on consortia and bulk buying will create a marketplace where small telcos can't compete, so the big boys get a bigger slice of the pie.
Losers:
  1. Phone equipment vendors.  Since voice is being phased out, at least hosted voice will get a little E-Rate funding for the next 5 years.  Somehow I don't think phone systems are going to get into C2, so PBX funding stops now.  Juicy PBX maintenance contracts are a thing of the past.  Eventually hosted and premises-based voice will be on an even footing, but for now, E-Rate tilts the playing field in favor of hosted.
  2. CIPA rebels.  If you don't comply with CIPA, you can only get funds for telecommunications.  As voice gets phased out, there goes most of your funding.  If you have multiple sites and you connect your buildings with circuits purchased from a company that files a 499, then you should still be able to get funding for those circuits, but that's it.  Libraries and private schools get left out again.
  3. Cell phone companies.  Just think: 90% applicants are going to find that their cell phone costs triple next year (from 10% of the actual cost to 30%).  Suddenly, giving employees a stipend to use their own phones looks good.  On the other hand, it sounds like the FCC does not intend to toss mobile data out of the program, so maybe the cell phone companies will be able to sell that to all the schools that can't get C2 funding for Wi-Fi.
  4. Schools serving special populations.  $150/student may be close to enough if you have a 15:1 student-teacher ratio, but when the ratio is 4:1, and you add in therapists, etc., and your student population is small, it's a sad joke.
  5. Schools that need Wi-Fi now.  These changes don't take effect until July 1, 2015, and even then, only applicants with extremely high discounts will get Wi-Fi funding.  And all schools need Wi-Fi now; schools need to prepare for all their students to take PARCC or SBAC in 8 months.  The FCC arrives with a couple of buckets 3 months after the house burned down.

Who'll get C2?

It would irresponsible to speculate on who will get C2 funding before the Seventh Report & Order (7R&O) has even been finalized.  So here I go.  I took enrollments from Funds for Learning and made the table below.
Disc. Enrollment Pre-discount New disc. Funding Cumulative
90% 4,001,855 $ 600,278,250 85% $ 510,236,513 $    510,236,513
89% 801,583 $ 120,237,450 85% $ 102,201,833 $    612,438,345
88% 1,095,160 $ 164,274,000 85% $ 139,632,900 $    752,071,245
87% 1,043,237 $ 156,485,550 85% $ 133,012,718 $    885,083,963
86% 953,147 $ 142,972,050 80% $ 114,377,640 $    999,461,603
85% 1,652,725 $ 247,908,750 80% $ 198,327,000 $ 1,197,788,603
84% 1,397,768 $ 209,665,200 80% $ 167,732,160 $ 1,365,520,763
83% 2,583,654 $ 387,548,100 80% $ 310,038,480 $ 1,675,559,243
82% 1,410,271 $ 211,540,650 80% $ 169,232,520 $ 1,844,791,763
81% 1,332,516 $ 199,877,400 80% $ 159,901,920 $ 2,004,693,683
80% 2,586,563 $ 387,984,450 80% $ 310,387,560 $ 2,315,081,243

What is that table?  The first two columns show the total enrollments of schools and districts that applied at each discount level in FY 2014.  For pre-discount, I multiplied enrollment by $150.  The new discount is my guess at what the discount will be under the new district-wide discounts and top discount.  Using the pre-discount and new discount, I calculated how much funding will be needed to cover each discount level.  Then the cumulative column shows how much funding we'll need to cover all the schools at that level and higher.

So by looking at that last column, we can see how much funding we'll need.  So if everyone applies for C2, and everyone is successful, $1 billion will cover schools that were at 86% and higher last year.  To cover all the way down to 80%, we'd need $2.3 billion.

So over the course of the next two years, the $2 billion that the Chairman leveraged out of the fund will give $150/student pre-discount to almost all the school districts with more than 50% of their kids eligible for free or reduced lunch.

Does your district have less than 50% NSLP participation?  No C2 for you.

Who's doing the letting?

I think I'll start the day with a petty rant.

You know what bugged me most about Chairman Wheeler's statement yesterday?  He said, "last year, no E-rate money at all was available for Wi-Fi services or other internal connections."  Live, he added emphasis to the point, saying something like "Zero. Zip."

Why was there no funding?  Because the Chairman chose to violate his own rules.  The rules at 47 C.F.R. 54.507(g)(iv) are pretty clear: he should have taken whatever funding was available ($200 million at the time, I think) and given 90% applicants a pro-rated share.

Actually, the Chairman announced that he'd found an extra $2 billion just before the decision was made to deny all 2013-2014 P2 requests, so he could have rolled $1.5 billion of that into 2013-2014 and funded P2 for all the 90% applicants.

So when he says, "the program has let students, teachers and local communities down" by not funding P2, I hope he realizes that he chose to let them down.

Friday, July 11, 2014

7R&O bits

So what is going to be in the Seventh Report & Order?  Responsible actors will wait until the text is released, but this blog is my opportunity to be irresponsible, so here is what I heard from the public meeting.
1.  2-year “trial” of Wi-Fi chimera
2.  85% top discount for C2
3.  Pricing transparency [whatever that means]
4.  Encourage consortia
5.  Lowest Corresponding Price [I can't wait to see how]
6.  Streamlined application
7.  Electronic filing required
8.  District-wide discounts
9.  New definition of rural
10. Direct invoicing by districts [I assume this means BEAR checks straight from USAC to applicants]
From Commissioner Clyburn:
11. $2.30/sq.ft. for libraries C2
12. The floor for small facilities will be increased (above $6,000)
13. Category 2 will not be prioritized over Category 1
Commissioner Rosenworcel:
14. Goals: 100 Mbps/1,000 students to all schools near-term, eventually 1 Gbps/1,000 students; library goals TBA
15. Collect better data
16. All C1 will be honored before anyone get C2
17. Raising cap in an FNPRM
Commissioner Pai:
18. 158-page order
19. No technology planning
20. Multi-year contract streamlining only for fixed-price five-year contracts
21. Preferred contracts must be considered
22. Applicants must appeal to USAC before appealing to FCC
23. Managed Wi-Fi eligible
24. C2 funding will be a five-year budget
25. Cap increase coming in December
26. Got final version of order at 10:13 a.m.
27. Pre-discount cost of Wi-Fi increased by $300 million overnight
Commissioner O’Rielly
28. It should be $/toilet, not $/sq.ft. for libraries.  OK, that's not really what he said.  But I never pass up a chance to put "toilet" in this blog.
John Wilkins:
29. Per-student funding goes away in 2 years, Wi-Fi remains eligible after that 
30. $1 billion dollar in non-broadband: save $350 million year 1
31. Projecting 10-20% efficiency savings


Other interesting things I heard:
  • Commissioner Pai called for removing competitive bidding rules, while Commissioner O'Rielly decried the changes in the order that loosened competitive bidding requirements.  I don't expect the Republicans in lock-step, but it's surprising to see them on opposite sides of the Chairman.
  • Chairman Wheeler said: “Money which is not spent does not exist.”  My bank assures me that the money in my savings account does exist.  I hate this idea that savings is waste and prudence is irresponsible.  The $2 billion that's going to Wi-Fi means that USAC will have committed $2 billion more than it has.
  • Chairman Wheeler said: “Because of what we do today, 10 million kids will be connected who otherwise wouldn’t.”  Sorry, but $1 billion in funding means under $1.2 billion pre-discount at 85%, which only covers 7.8 million kids.  And that assumes that C1 demand stays under the cap, which it won't.  And since most of those 85%-discount applicants have been getting P2 all along, they have Wi-Fi now; the C2 money is just to get those 7.8 million kids better Wi-Fi.
  • Chairman Wheeler said something about being able to see how our reforms will affect funding demand before deciding on whether to increase the fund.  We won't know how these reforms affect demand until USAC releases demand numbers, probably in April 2015.  Does that mean the Chairman plans to wait until then to start talking about raising the cap?
More when the actual order comes out.

Which states win?

Sen. Ayotte's complaint in a recent letter than NH was paying out more in E-Rate dollars got me wondering: what states pay the most in, and which get the most out?  Fortunately, the FCC supplies this info to Congress, so I checked out their report on 2012 data.  I put it into a spreadsheet to play with.  Yup, NH is down there, but it's not quite true that "New Hampshire ranks 50 out of 50 states when it comes to return on our E-Rate dollar."  It's true that NH put 4 times as much into the program as it gets out, but Hawaii put in more than 5 times as much as they got out.  So NH is 49 out of 50.

Who are the big winners?  Alaska, of course, getting over 7 times as much out as they put in.  Oklahoma and New Mexico are the only other states getting more than twice what they put in.

California and Texas are #1 and #2 in both dollars put into the fund and dollars taken out.  Both put in more than they get out, and CA has the highest net flow into the program.  Which state has the largest positive cash flow?  Oklahoma.

I also looked at per capita cash flow.  Who wins? Alaska, getting $406.87 per 1,000 residents.  ND comes in second, getting $112.96.  Last place?  DC, which loses $42.90 per 1,000 residents to the E-Rate.  So I guess before Commissioner Pai complained about how much funding DC was getting in spite of not being rural, he might have looked at whether DC was a net winner or loser.  NH, by the way, is 45th out of 50 in per-capita loss.

Thursday, July 10, 2014

Bipartisan agreement on one thing

Congress is weighing in on the E-Rate.  Here's a summary of letters I've seen.

Senators Rockefeller and Markey, who call themselves (with some justification) "the founders of the E-Rate," wrote a letter on Tuesday with a few criticisms:
  1. "It would be ill-advised to guarantee a permanent set-aside for Wi-Fi....  All basic Internet connectivity requests...should be honored before Wi-Fi funding is made available."  
  2. "We are opposed to...per student or per square feet distribution mechanisms for any aspect of E-Rate..."
  3. "The permanent funding cap for the E-Rate program needs to be raised."
I pretty much agree with the founders of the E-Rate.  They even use a capital "R" in "E-Rate"!

Senator Thune released a statement yesterday saying:
  1. "Chairman Wheeler ... cannot realistically expect to pay for [his proposal] without forcing Americans to pay more for communications services or diverting E-Rate funds that support necessary connectivity...."
  2. "If Chairman Wheeler is unable to move forward with reforms that have the bipartisan support of his FCC colleagues, he should postpone Friday’s scheduled vote...."
I agree with #1.  Most educators and lawwmakers would agree that voice connectivity is necessary for schools, but all the proposals I've seen divert funding away from that necessary connectivity.  As for #2, bipartisanship is so last century.  I'd like to see the Chairman turn each major part of the reforms into a separate order, and then vote on each of them separately.  It would be fun to see how that plays out.

Representatives Upton and Walden wrote today, with different criticisms:
  1. "Recent reports raise concerns about your commitment to work within existing funds."  
  2. "We are also troubled by press reports that you have promised to increase the E-Rate budget...."
I agree with the representatives' use of the capital "R" in "E-Rate," but otherwise....  When did the Chairman make a commitment to work within existing funds?  I keep hearing him say he'll increase the fund if warranted.  But I haven't heard him say it's warranted.

Senators Sabenow and Levin wrote today to say:
  1. "We oppose...a per student or per square feet calculation for Wi-Fi."
  2. "We also oppose changing the program's poverty calculation from school-based to district-based...."
I agree with #1, but I think the district-level discount is better; it can eliminate a lot of small complexities in the program.

If both sides of the aisle dislike the Chairman's proposal, does that mean it's a good compromise? 

What one principle unites all the congresspeople?  "E-Rate" with a big "R"!

This Pai sticks in the Chairman's craw

Commissioner Pai isn't letting up.  He's released a financial projection of what will happen if the Chairman's reform proposal is implemented.  I'm not clear on where the numbers come from.  But I'm guessing it's based on projected C1 demand (don't call it P1, because it no longer includes a lot of P1 services like phones and Web hosting), a fixed cap, and $1 billion/year taken out of the fund for the Wi-Fi chimera.

The good news: the Commissioner seems to no longer believe in the red-tape funding gap mirage.  When he was doing the calculations for his proposal, that was good for $400 million in savings each year.

I don't think it's fair to say that the plan will "cut" funding.  Even if the Commissioner's pessimistic numbers were right, it's not that anyone is cutting funding, it's that funding for Internet is being crowded out by other priorities.  Voice services, webhosting, email, etc. are being cut (with Commissioner Pai's support).

Yesterday, Commissioner Pai said that the FCC plans to increase the contribution factor after the elections.  Seems to me that you can't have it both ways: the Chairman may be planning to jack our phone bills, or he may be starve the fund so that Internet access won't be funded, but he can't be planning to do both.

But I have to say, I agree with Commissioner Pai's underlying point (and so do a lot of schools and libraries): we should be told how the Chairman plans to pay for years 3 through 5 of his plan.  Commissioner Pai's not the only one worried that come Year 3, a lot of applicants are going to left high and dry.

What's up with the emphasis on rural applicants?  Commissioner Pai always seems to be focused on the rural.  In his reform proposal, rural applicants got twice as much funding as non-rural.  And in this press release, he says, "many rural schools would lose all of their funding."  Actually, the non-rural schools would be first in line, since rural schools get a 10% goose in their discount at lower levels.  So the first to lose funding would be non-rural schools with 0%-1% low-income kids (20% discount).  Then the rural schools with 0%-1% low-income kids (25% discount).  Then non-rural with 1%-19.5% low-income (40%).  The 50% band has both rural schools with 1%-19.5% low-income and non-rural with 19.5%-34.5% low-income.

So while the cuts imagined by Commissioner Pai would affect rural school children, they would affect a lot more non-rural children.

So Commissioner Pai gets a tip of my hat for pointing out that the emperor loses his clothes in year 3 of this reform, and a wag of my finger for overstating the harm, especially the harm to rural students.

Wednesday, July 09, 2014

Pai in your face

Looks like Commissioner Pai has had enough.  He released a statement saying that the Chairman's office "rejected almost every suggestion that I made" about E-Rate reform "and is determined to pass this item on a party-line vote."  In other words, the Chairman is going to go with the proposals that have the support of the majority of the Commissioners, and will not be adopting proposals that are supported by a minority of Commissioners.

I just wish Commissioner Pai had told us what suggestions he made, and which of them were accepted, which rejected.  Seems to me that if you're going to complain publicly, you ought to tell us a little more about what's got you complaining.

I agree with Commissioner Pai on complexity: the Chairman's reform proposal adds new layers of complexity to the program.  Not only does the E-Rate rules hydra get more heads, but he's introduced a chimera and a minotaur.  I never thought the program would get simpler, but I'm surprised at how much more complex it's getting.

Commissioner Pai does offer an explanation for why increasing the cap is not part of the Chairman's reform proposal: the Chairman is waiting until after the November elections to raise the contribution factor.  That seems likely to me.  And given the choice between angering the applicant community and giving Republicans a campaign issue, I'd say the Chairman will choose to anger applicants, so the cap is not going up, even if ISTE gets 15,000 signatures.

Tech plan smackdown

Wow!  I was just searching the FCC site for recent comments on E-Rate reform, and discovered that since Monday, 225 comments have been filed!  It seems that it was a bit of a miscalculation for the Chairman to have released his proposal shortly before the ISTE and ALA conferences.  My unscientific poll shows many notes from schools (including a petition with 1500 signatures from ISTE) asking to have the funding cap raised, and from libraries asking for the library's C2 allowance to be increased.  And a few statements of support for the Chairman's proposals.

Oooh, look!  A little spat between SECA and SETDA.  The State E-Rate Coordinators Alliance (SECA) proposed scrapping the tech plan requirement, and that made it into the Chairman's proposal.  The State Educational Technology Directors Association (SETDA) thinks tech plans are good, and said so in a recent comment.  SECA fired back.  C'mon, SETDA, you've still got a day to respond!

Who's right?  Well, maybe I'll just side with whichever one spells "E-Rate" correctly.  Dang: both wrote "E-rate."  I guess I'll have to decide based on the merits.

SECA, of course.  I've advocated tossing the tech plans since 2007.  Heck, the FCC has been waiving tech plan requirements since 2006.

Tuesday, July 08, 2014

How do libraries make out?

Is $1/square foot a fair amount of Category II (C2) funding for libraries?  Well, I can't really talk about "fair," since that's in the eye of the beholder, but let's compare the funding of a few libraries with the funding for the schools in the same area.  How will I decide which applicants to include in my study?  The same way I always do: I'll take what's easy to find.  My sample is selected because the library square footage popped up on Google.  I got the square footage numbers off the Internet, so I know they're true and didn't bother to verify them.  Also, I matched the school district to the library just based on name; I can't swear the service footprint is identical.  Where I noticed a small library (including bookmobiles), I boosted funding up to the $6,000 minimum, but I didn't do that for any schools (because the difference would have been negligible).  So the data below is a blurry keyhole view of the landscape.
Community Enrollment School $ Library $ Ratio
Holyoke, MA 5,316 $797,400 $40,000 19.9
Chandler, AZ 42,876 $6,431,400 $124,500 51.7
Denver, CO 70,717 $10,607,550 $752,137 14.1
Gilpin County, CO 434 $65,100 $12,000 5.4
Taylor County, FL 3,040 $456,000 $7,158 63.7
Broward County, FL 222,527 $33,379,050 $1,447,408 23.1

The last column shows the multiple of funding that the school district gets.  What is a fair multiple?  Well, you're not likely to think that both 14:1 and 51:1 are both fair, so everyone will find that unfair in at least a couple of cases.  So no matter what you think is fair, not every library will get a fair amount of funding.

Taking a step back, the real funding problem for libraries is not going to be the $1/sq.ft. formula.  Most libraries don't apply for the E-Rate, and even fewer are going to apply for C2 (since it requires CIPA compliance).

It gets even worse: now that they're phasing out voice service, a single-site library that's not CIPA-compliant is eligible for $0 in E-Rate funding.  The only service left in the program that does not require CIPA compliance is digital transmission service (DTS) purchased from a telecommunications carrier.  Single-site applicants generally don't need DTS, so they'll get bupkus.

Take that, rural libraries.

The Chimera takes shape

We're getting more news about the Wi-Fi-Rate chimera, the separate fund-within-a-fund that the FCC seems about to create.  Based on statements from the FCC, the plan for Category 2 (C2, née Priority 2) funding is something like:

  1. Each applicant location will get a set amount of funding for Wi-Fi:
    1. Schools will get $150/student
    2. Libraries will get $1/sq.ft.
    3. Small applicants will get a minimum of $6,000.
  2. The top discount will be 80%.
  3. Applicants will be approved once every 5 years.
  4. Applicants who are approved will be allowed to use the funding any time over the course of 5 years.
  5. Equipment purchased will be assigned to a location, and cannot be moved from there.
And what do I think of that?

The good:
#2 is a start.  Let's cut the top discount rate to 65% for everything, like they did over at the Rural Health Care program, and for the same reasons.

The bad:
#3 is doubling down on a bad rule.  The 2-in-5 Rule was an utter failure, and drove up applicants costs.  The 1-in-5 will be worse.  Well, except #4 helps there.  5 years is too long to wait to refresh wireless access points, but this whole plan is about giving applicants less than they need, so forcing them to keep outmoded access points isn't the worst of it.

#5 is a nightmare.  How many schools open each year?  How many schools close?  What happens to them?  And what about schools that decrease or increase enrollment?  And given that the discount level is set at the district level, there is no reason for an applicant to be transferring equipment between locations, other than a change in need.  Bad, bad rule.

The ugly:
#4 is a kind of good idea with a bad outcome.  The FCC is saying: "Apply for C2 funding any time.  When you get funded, you will have 5 years to spend the money."  What I will be saying to clients: "If you intend to install equipment any time in the next 5 years, apply this year and keep applying until you get funded."  Why would anyone wait to apply?  Get the funding now, and figure out how you're going to spend it later.

#1 is a so-so idea (per-student funding) gone horribly wrong.  The idea that the funding will be allocated per building is just so wrong.  First, not every building needs the same amount of infrastructure.  Think there will be as much BYOD in an elementary school as in a high school?  Second, in many districts, the main Internet connection is in an administrative building, but there will be no way to get C2 funding to buildings without students.  If the main Internet connection isn't in an admin building, it will be in one of the schools, which means that school will have higher C2 costs, without getting more C2 funding.  Third, in order to keep districts from moving equipment after installation, you need to have rule #5, and I just pointed out the problem there.

Of course, the goal of this chimera is not to deliver what applicants need, but to keep costs down, so the fact that admin buildings and high schools will be underfunded in most districts is only sort of a problem, since the district as a whole will be underfunded.

Thursday, July 03, 2014

Putting the math to work

Over at Funds for Learning, John Harrington has looked at the numbers and decided that the math for the Wi-Fi chimera works.  As usual, FFL is not wrong, but I don't agree with the analysis.  The math does work if the $150/student that we've been hearing about will be spread over five years.  So if every school gets $30/student/year, it will cost about $1 billion per year, just like the Chairman plans.

On to the problems.
  1. The easy one: libraries.  If it will cost $1 billion/year to fund schools, how do we fund the libraries?  We're going to need a bigger fund.
  2. $30/student/year?  ISTE clearly thinks it's going to be $150/student once every 5 years.  I'll do that math later.
  3. The discount levels are a-changin'.  The rumor is that the Chairman's promise to "simplify discount calculations" means school districts will have a single discount level for all locations, based on the enrollment and NSLP totals for the district.  No more 86% districts.  How will this affect funding?  I don't know, but I'd guess that overall, it will lower funding demands; I suspect that 86% districts are more likely to move to 80% than 90%.
  4. Who's getting their discount cut?  The last two columns show 1) funding at current discount rates, and 2) funding if all discount rates are cut 10%.  The only talk I've heard of cutting discounts is to make the top discount 80%.
Ooh look!  There's a table showing the number of students at each discount band.  I just can't resist numbers, so I dumped the data into a spreadsheet to play with.  You can download my spreadsheet and play, too.

I tried to get us to the new discount bands. I made some assumptions about rounding: applicants at 29% or below went down to 20%, 30% and above went up to 40%.  Beyond that, I rounded down on the 6, up on the 7, so 46% and lower became 40%, 47% and above became 50%, and so on.

I created columns for: 1) no change in discount, 2) everyone gets a 10% haircut, and 3) the matrix is the same, except the top discount is 80%.

I also created cumulative columns, so I could quickly see how much we'd need to cover all applicants at that discount rate and higher.

Looking at $30/student/year (if you downloaded the spreadsheet and are playing along at home, you can adjust the per-student amount in cell B2) with only the 90% group cut by 10%, the total demand is: $1.035 billion.  So we're short by $35 million plus whatever the libraries need.

Looking at the one-time $150/student:
  1. If the only change in discount levels is to reduce the top discount to 80%, then we'll need $2.67 billion just to cover the 80% applicants.  
  2. If everyone takes a 10% haircut, then with $1 billion we can probably fund all the 80% applicants (those would be the applicants who currently have a discount between 86% and 90%).  But we can't fund the 70% applicants (those with a current discount of (77% to 86%).  In fact, we'll need $1.8 billion just to cover the 70% applicants, so we can't fit them all in one year.  We'll need just over $1 billion to cover the 60% applicants.  After that it gets easier.
The math doesn't work using the current practice of denying all applicants in a discount band if there isn't enough money left to fund everyone in that band.  The FCC could always follow the rules, and pro-rate, but the howls would be considerable, especially if the 1-in-5 Rule is adopted.

The solution is to do what they should have been doing all along: if there isn't enough money left to fully fund the applicants in a discount band, go to the raw NSLP data.  Not enough funding to meet demand from 90% applicants?  Fund those applicants with 100% low-income kids, then the applicants with 99% low-income kids, and so on until the funding runs out.

The math shows me some challenges ahead.  We can't really say exactly what the challenges will be until we've seen the new rules, but one thing's clear to me: we're going to need a bigger fund.

With supporters like this...

I read the Chairman's recent blog post about his visit to the Acoma Learning Center in New Mexico, and I thought, "How hard did he have to search to find a rural library that received E-Rate funding?"  Then I look at the Learning Center's funding history.

Priority One Priority Two
FY Requested Approved Disbursed Requested Approved Disbursed
2014 $ 31,763.66 $ - $ -
2013 $ 10,668.43 $ - $ - $ 35,732.99 $ - $ -
2012 $ 35,386.53 $ - $ -
2011
2010 $ 73,658.73 $ - $ - $ 502,707.89 $ 477,724.33 $ 97,033.38
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000 $ 4,455.00 $ 4,455.00 $ - $ 292.50 $ 157.50 $ -
1999
1998

All those empty boxes?  Years in which the Learning Center didn't apply.

So when the Chairman said, "The FCC’s E-Rate program has supported basic Internet access for the Acoma Learning Center," I guess he's referring to that one equipment purchase in FY 2010-2011.  Otherwise, the center's gotten nothing.  They are a 90% applicant, so the only thing that's been holding them back from getting telephone, Internet, internal connections and basic maintenance funding is the application process and program requirements.

The Chairman did say the program needs to do more.  In his position, I might have described it as a failure of the E-Rate program, noting the old rules weren't working for rural libraries.

But here's another way to look at it.  The Learning Center's P1 need is about $10,000/year.  Under the proposed rules, a tiny library is going to get $6,000 in C2 funding once in 5 years.  So while Acoma only got funding once under the old rules, they got as much funding in that one year as they'll get in 9 years under the new rules.

Will the new rules work better to get rural libraries into the program?  I don't know, because we haven't seen them.  Here's what would work for Acoma: 1) remove CIPA, and 2) simplify the application process.  The first isn't really in the FCC's control.  As for the second, we've been told that there will be a simplified application process, but it seems to be only for low-dollar applicants whose prices fall below a certain threshold.  What are the chances that a remote library will get prices low enough to qualify for the simplified application process?  Not good.  Does $10,000 meet the "low dollar" threshold?  Maybe.  But I don't see how they'll stay low-dollar if they pull in some fiber to get a 100 Mbps connection.

So I don't expect the upcoming reforms to make the table above look better in the future.  How could we set them on the path to funding?  They could hire a consultant.  I know some people find E-Rate consultants distasteful, and I try not to promote the use of consultants too much in this blog, but this case is crying out for one.  I mean, if you saw that a company hadn't filed its taxes correctly a single time in 17 years, mightn't you suggest they contract with an accounting firm?

Wednesday, July 02, 2014

Modernizing the Facts

Let's take a look at the FCC's analysis of the effect of the proposed reforms.  Of course, my analysis is hampered by not knowing what the reforms actually are, but you know me, always happy to give my opinion without having all the facts.

"get high-speed Internet to all classrooms and libraries by 2019."  Wait, the plan is to cover all buildings in 5 funding years, starting in 2015-2016.  That means the last classrooms will be covered in 2020-2021.

"only 43 percent of public school districts report that their internal connections are capable of supporting a one device...per student...."  Almost accurate.  What 43% of tech directors said was that a typical school could support 1:1.  Not all the schools in their district.

"50 percent of schools lack the internal wiring...."  Just plain wrong.  The survey results showed that 50% of schools have Cat5 cable, which can support for 1000BaseT but is not preferred.  That same survey says that 77% of buildings have Cat5e.  So at most, 23% of school lack wiring to get them to gigabit speeds.  And some of those 23% may have Cat6 (which that survey says is in 61% of buildings).  So you can say "50% of buildings have Cat5 wiring, which is not preferred for gigabit speeds."  That would be deceptive, since most of those buildings also have Cat5e and Cat6 (and fiber, though presumably not to the desktop), but at least it wouldn't be incorrect.

"In Funding Year 2013, no funding was available for Priority 2 requests."  Not true.  About $200 million was available, but that was nowhere near enough to fund all the requests from 90% applicants, so rather than follow the rules, which say applicants at that discount level should have received a pro-rated portion of the remaining funding, the FCC followed tradition (and common sense) and didn't fund anyone in that band.  Mind you, Chairman Wheeler had just announced his plan to leverage an extra $2 billion out of the fund, so he could have opted to use some of that to fund P2 for 2013.  For that matter, since they were already ignoring the rules, they could have given all the 90% applicants a per-student allocation of, say, $150.

The graphs on page 3 certainly show the abject failure of the 2-in-5 Rule.  But where did they get those library numbers?   I spot-checked the number of library locations on applications in 2012.  I found 78  471s where the applicant was a library.  4 of those applicants had 2 applications that year, so I took a look at 74 unique 471s.  The total number of entities listed on the Block 4 of those applications is 362.  Not 177.  I'd want to double-check my number before I published it in a report, but I'm sure that 177 is way off.  Also the majority of libraries choose not to apply for P1 funding, so it seems unfair to use the total number of libraries for the big bar every year.

How much will we save by tossing those services listed on page 4?  (Hint: it's not the $1.2 billion claimed.)  Let's look at USAC's most recent analysis. Pagers: $0.001 billion.  Email: $0.01 billion.  Voice: $0.57 billion ($0.55 billion if you take VoIP out).  What else can we toss in there?  They didn't mention webhosting, but I'm sure that's on the chopping block: $0.027 billion.  Let's toss voicemail: $0.0002 billion.  The Chairman suggested chucking DNS hosting: oops, too small to be listed.  I'm not even halfway to $1.2 billion.  And those numbers are approvals, not disbursements, so the actual savings will be even smaller.  Plus, if you leave VoIP in the program, then the voice dollars will just flow to VoIP, and the savings will be less than $0.03 billion.

The table starting on page 7 is false.  The school column is just optimistic, assuming that every school that was on a 471 last year will be getting Wi-fi funding.  It's true that setting aside for Wi-Fi, then doling out a certain amount per student, you can theoretically give Wi-Fi funding to all those schools.  Not as much funding as they need, but some funding.  But there are 132,270 schools out there, so even if you give a little money to the 116,000 schools that applied for funding last year, you're still missing 16,000 schools.  You won't get to 99% of students if you leave out 12% of the schools.

But where the table is flat-out false is the "Additional Libraries" column.  It assumes that just because Wi-Fi funding becomes available, all libraries will get some.  Wrong for two reasons.
  1. Only about 40% of public libraries apply for funding.  The majority of libraries just leave E-Rate money on the table.  And I expect that the percentage of libraries applying will drop when voice is thrown out of the program, because that's the only funding some libraries get.  
  2. CIPA.  Many librarians in this country value our constitutional rights more than a little federal funding, so 29% of libraries don't apply for E-Rate because of CIPA.  That's not going to change just because the FCC's going to toss them a one-time bone of a few thousand to get Wi-Fi. 
Wow, I did not expect this to turn into a fact-checking mission, but there were so many glaring errors.

Tuesday, July 01, 2014

That's not what I was asking

The FCC seems to have noticed that applicants are very concerned about some of the rumors the reform proposal now circulating in their offices.  The obvious solution: publish the proposal.  The FCC solution: release a Q&A blog post -- long on spin, short on specifics -- and a report on the effect of the reform.  Let's take a look at the blog post first.

First, the Q&A:
1. How do the numbers add up?  
The FCC says: we've already leveraged the fund to get $2 billion for the next 2 years.  After that, the funding will come from: 1) $1.2 billion in savings from tossing voice, etc. out of the program, and 2) cost savings.
I say: 1) Sorry, but the services that the Chairman talked about chopping gets us less than $0.7 billion.  And most of that evaporates if you leave VoIP in the program, because applicants will just move their voice to IP.  2) Transparency will help applicants lower costs by maybe 10%, which is nowhere near enough.  The GSA minotaur isn't going to save money.  Neither are consortia.

2. What about Priority 1?
The FCC says: 1) Price transparency will allow rural applicants to use pricing for similar applicants to negotiate lower pricing.  2) Low-dollar applicants will not need to go through competitive bidding if they meet certain price thresholds.
I say: 1) How does that work?  A rural library in Colorado can go to CenturyLink and say, "Hey, libraries in Kansas are getting a lower rate from AT&T than you're giving us here."  And CenturyLink will respond, "You're right; you should move to Kansas."  Price transparency is helpful where there is competition.  Which is not rural America.  2) That sounds like a great way to reduce administrative overhead, but it doesn't really lower costs, does it?

3. What does this mean for rural schools?
The FCC says: 1) Wi-Fi funding for rural schools will be increased by 75%.  2) "substantially easier for low dollar purchasers, including smaller, rural schools and libraries to purchase 100 Mbps or higher business/enterprise service commonly available below a certain price."  3) "greater pricing transparency in the program would help lower rates for services in rural areas."
I say: 1) Yup, by mortgaging the program to come up with $2 billion, and reducing the non-Wi-Fi services available, Wi-Fi funding goes up.  Since no one knows what portion of Internal Connections goes to Wi-Fi equipment, the 75% growth can't be disproven.  2) Chop!  There goes another head off the hydra.  First new head: I have to hope that someone offers service below the cutoff price, which is much less likely in a rural area.  Second new head: I have to figure out if I need enterprise-class service, or will my cable company's "business class" cable-modem service meet this definition?  Third new head: I guess if the service provider says "100 Mbps," that's good enough, or do I need to show that I'm actually getting 100 Mbps?  Fourth new head: If the price goes up later, do I have to bid it out then?  3) As stated above, transparency will lower rates only in areas where there is competition, and where applicants are close enough that comparisons can be made.

4. What does this mean for urban schools?
The FCC says: 60% more Wi-Fi funding.
I say: See the first item in the question above.  Let's see: the FCC is mortgaging the future of the program to temporarily increase the fund by less than 30%, and somehow that's going to increase funding for everybody by 60-72%.  Show us that math.

5.  Does the proposal make E-Rate funding more equitable?
The FCC says: The proposal "would require a common sense budget for Wi-Fi spending...."  The current system only benefits a few applicants.  The reform will spread the money but "that those who need it most receive the most."
I say: If by "common sense" you mean "based on a gut feeling, not actual data," then I agree.  The current system benefits only a few applicants because: 1) 90% is too close to free, and 2) there isn't enough money in the fund.  The proposal addresses both issues, though I think 80% is still too close to free and there still isn't enough money (hence the need for the 1-in-5 nonsense).  Based on the stats above and mention of "bump," it sounds like rural schools will get more.  Is that equitable?  Depends who you ask.

"What our discussions have revealed is that everyone shares the same goals...."  Only if you make the goals really vague like "modernize the program."  When it gets down to concrete steps, the one that has the strongest support is to increase the size of the fund.  And that isn't in the proposal.