Voglio sapere a riguardo di…

40753

Federal Register / Vol. 86, No. 142 / Wednesday, July 28, 2021 / Rules and Regulations TABLE 6PROVIDER MATCHES TO GTL BY FACILITY TYPE
Smaller Jail Number of matches Securus
ICSolutions
CPC
NCIC
Legacy
Pay Tel
CenturyLink
Correct
ATN
Crown
Prodigy

Larger Jail Number of matches
Percent
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED

REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED

20. The numbers in parentheses represent the percentage of all non-outlier and nonGTL contracts that each firm has, thereby allowing for a comparison of the frequency of nearest neighbor matches to the overall frequency in the data. Unsurprisingly, given the large market share of each, Securus is a frequent match to GTL. Of the REDACTED
GTL contracts included in the analysis, REDACTED of them 19.7% include zero Securus contracts in their neighborhood;
REDACTED 34.8% include one Securus contract in their neighborhood; REDACTED
29.4% include two Securus contracts in their neighborhood; and REDACTED
16.1% include three Securus contracts in their neighborhood. By neighborhood, the Commission refers to the set of three matched contracts for each GTL contract. On average, a GTL contracts neighborhood is comprised of REDACTED 47.3% Securus contracts.
As Securus comprises roughly 40% of all non-GTL contracts in the data, the results are reasonable and suggest that Securus does not
Prison Number of matches
Percent
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED

REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED

Percent
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED

have an outsized influence on the matching relative to its size in the market. After Securus, the providers whose contracts constitute the largest number of neighbors to GTL contracts are ICSolutions, CPC, and NCIC, with the average neighborhood consisting of REDACTED contracts from each provider, respectively.
21. That no firm plays an outsized role in the nearest neighbor matching holds across the different types of facilities. REDACTED
In general, the smallest firms in the market tend to be under-represented in the matching, likely because scale economies make the bigger players look more similar along multiple dimensions of a contract, even within a particular facility type.
22. The results of this analysis indicate that GTL is being matched to every other firm in the data at least some of the time. Though its nearest neighbors are usually other large providers, that is in no way surprising. The variation in the match data supports the validity of the results, while shedding
Overall
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED

Number of matches REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED

Percent REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED
REDACTED

additional light on the contracts that look closest to GTLs for the purposes of the data adjustment procedure.
3. Determining Lower Bound for Interim Rate Caps 23. With confidence that the outlier and GTL data adjustment procedures are valid and robust to a variety of assumptions, the Commission can now construct the lower bounds for the zones of reasonableness. As with the upper bound approach, the Commission defines the lower bound as the mean plus one standard deviation of perminute contract costs, separately for prisons and larger jails. These estimates rely on the full contract-level data excluding the identified outliers and replacing the original GTL cost data with the per-minute cost estimates derived from the nearest neighbor adjustment procedure. The Commission presents the lower bound estimates in Table 7 below.

Table 7 - Lower Bound Estimates Lower Bound-Weighted GTL Adjustment of Std. Dev.
Contracts $

Larger Jails
of Contracts 179

Prisons
129

24. As with the previous results, the Commission presents lower bound estimates derived from a weighted average GTL
adjustment as well as more conservative estimates based on the maximum of GTLs nearest neighbors. As both approaches are valid, the Commission selects the weighted average results as the estimates of the lower bound for the zone of reasonableness. For prisons, the lower bound is $0.064, and for larger jails, the lower bound is $0.08. These
VerDate Sep<11>2014

19:16 Jul 27, 2021

Jkt 253001

Mean
Std. Dev
$

$

0.070

0.108

0.058

0.088

are the most plausible, lowest estimates of per-minute interim rate caps across all contracts in the data.
4. Maximum GTL Costs Support the New Interim Provider-Related Rate Caps 25. The Commission has established the lower bounds of the zones of reasonableness as being $0.064 for prisons and $0.080 for larger jails based on an analysis that removes outlier observations and adjusts unreliable
PO 00000

Frm 00073

Fmt 4701

Sfmt 4700

GTL per-minute cost data. Given GTLs size and presence in the inmate calling services market, the Commission now determine the maximum per-minute costs that GTL could hypothetically incur that would still support the interim provider-related rate caps. That is, the Commission asks what GTLs highest average per-minute costs would need to be, separately for its prison and larger jail contacts, such that the overall per-minute cost plus one standard deviation across all
E:FRFM28JYR2.SGM

28JYR2

ER28JY21.003

khammond on DSKJM1Z7X2PROD with RULES2

Larger Jails