On April 1, 2019, the USCIS (U.S. Citizenship and Immigration Services) launched the H-1B Employer Data Hub to provide information on employers petitioning for H-1B workers. Also posted is a description of its use titled "Understanding Our H-1B Employer Data Hub". The hub page states that the complete files for individual fiscal years can be downloaded from the H-1B Employer Data Hub Files page. These files were used to create an R Shiny application that extends the current functionality of the USCIS Data Hub. Following is an example:
Go to the USCIS H-1B Employer Data Hub, select 2020 for the Fiscal Year, and press the Search button. The following first 10 of a reported 55,239 entries will be displayed:
Fiscal Initial Initial Year Employer Approval Denial 2020 0 1 2020 0 0 2020 1 0 2020 01INTERACTIVE INC 1 1 2020 01INTERACTIVE INC 0 1 2020 05MEDIA INC 1 0 2020 0965688 BC LTD DBA PROCOGIA 1 1 2020 1 800 FLOWERS COM INC 1 0 2020 1 BETHESDA DRIVE OPERATING CO LLC 0 0 2020 1 DUST TECHNOLOGY LLC 0 1Additional information is available by clicking the arrows on the far right. Now, go to the R Shiny app, ensure that the Start Year and End Year are set to 2020, place checks in the "Output all records" and "Output employer details" checkboxes, select "Employer" in the "Sort by" select list, and press the Submit button. The first row of data output reports that there are 55,666 records, same as was reported by the data hub. Following are the first 10 rows in the last table output:
FYear Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R
1 2020 55239 122894 18276 303830 21639 426724 87.1 93.4 91.4
Employer Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R NAICS Tax.ID State City ZIP
1 1 1 0 0 0 1 100.0 NaN 100.0 54 777 CA GLENDORA 91740
2 1 0 0 1 0 1 NaN 100.0 100.0 62 458 CA EMERYVILLE 94608
3 1 0 1 0 0 0 0.0 NaN 0.0 71 3557 NY ASTORIA 11106
4 01INTERACTIVE INC 1 1 1 0 0 1 50.0 NaN 50.0 54 9852 CA CITY OF INDUSTRY 91745
5 01INTERACTIVE INC 1 0 1 0 0 0 0.0 NaN 0.0 54 9852 CA HACIENDA HEIGHTS 91745
6 05MEDIA INC 1 1 0 0 0 1 100.0 NaN 100.0 54 4558 CA SAN DIEGO 92130
7 0965688 BC LTD DBA PROCOGIA 1 1 1 1 0 2 50.0 100.0 66.7 54 209 WA BELLEVUE 98006
8 1 800 FLOWERS COM INC 1 1 0 3 0 4 100.0 100.0 100.0 45 7311 NY CARLE PLACE 11514
9 1 BETHESDA DRIVE OPERATING CO LLC 1 0 0 1 0 1 NaN 100.0 100.0 62 5614 NY HORNELL 14843
10 1 DUST TECHNOLOGY LLC 1 0 1 0 0 0 0.0 NaN 0.0 99 5330 MO CHESTERFIELD 63017
As can be seen, the rows are pretty much identical. The only difference is that the three lines with no employer are reversed in order. Also, note that lines 3, 5, and 10 have workers equal to zero. This is because there were no initial approvals (I.A.) or continuing approvals (C.A.) for those employers.
The column names I.A., I.D., C.A., and C.D. stand for Initial Approval, Initial Denial, Continuing Approval, and Continuing Denial, respectively. If the Initial Approval label is clicked in the header of the data hub output, the data will be sorted in descending order by Initial Approval and the following 10 rows will be output:
Fiscal Year Employer Initial Approval sorted descending. Select to sort ascending. Initial Denial Fiscal Initial Initial Year Employer Approval Denial 2020 AMAZON COM SERVICES LLC 3737 94 2020 INFOSYS LIMITED 3487 1958 2020 TATA CONSULTANCY SVCS LTD 2401 264 2020 COGNIZANT TECHNOLOGY SOLUTIONS US 1991 1519 2020 MICROSOFT CORPORATION 1788 42 2020 GOOGLE LLC 1673 21 2020 CAPGEMINI AMERICA INC 1399 357 2020 HCL AMERICA INC 1301 399 2020 IBM CORPORATION 1281 116 2020 FACEBOOK INC 1168 7If "I.A." is selected in the "Sort by" select list in the R Shiny app, following are the first 10 records output in the last group of data on the page:
Employer Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R NAICS Tax.ID State City ZIP
1 AMAZON COM SERVICES LLC 3 3737 94 7556 119 11293 97.5 98.4 98.1 45 4687 WA SEATTLE 98121
2 INFOSYS LIMITED 17 3493 1968 6237 1562 9730 64.0 80.0 73.4 54 235 TX RICHARDSON 75082
3 TATA CONSULTANCY SVCS LTD 7 2403 264 6008 673 8411 90.1 89.9 90.0 54 9806 MD ROCKVILLE 20850
4 COGNIZANT TECHNOLOGY SOLUTIONS US 7 1992 1520 12740 2493 14732 56.7 83.6 78.6 54 4155 TX COLLEGE STATION 77845
5 MICROSOFT CORPORATION 4 1788 42 5192 75 6980 97.7 98.6 98.4 51 4442 WA REDMOND 98052
6 GOOGLE LLC 1 1673 21 4129 35 5802 98.8 99.2 99.0 54 3581 CA MOUNTAIN VIEW 94043
7 CAPGEMINI AMERICA INC 2 1399 357 3775 169 5174 79.7 95.7 90.8 54 5929 IL CHICAGO 60606
8 HCL AMERICA INC 3 1302 399 2792 511 4094 76.5 84.5 81.8 54 5035 CA SUNNYVALE 94085
9 IBM CORPORATION 2 1281 116 1035 52 2316 91.7 95.2 93.2 54 1985 NC DURHAM 27709
10 FACEBOOK INC 1 1168 7 3196 25 4364 99.4 99.2 99.3 51 5019 CA MENLO PARK 94025
As can be seen, the data is identical to the data output by the USCIS data hub. It does include the additional column "Workers". By default, this is set to the sum of I.A. (Initial Approval) and C.A. (Coninuing Approval) but this can be changed via the "Include in Workers" checkboxes. In any case, these are the 10 records with the highest number of Initial H-1B approvals in 2020. However, there appears to be a separate record for each combination of Employer, NAICS, Tax.ID, State, City, and ZIP.
The R Shiny app will also show the totals by year and by employer. This can be seen in the following screenshot:
Following are the top 10 records from the prior section:
FYear Employer I.A. I.D. C.A. C.D. NAICS Tax.ID State City ZIP Workers 1 2018 AMAZON FULFILLMENT SERVICES INC 2392 22 1980 45 45 4687 WA SEATTLE 98121 4372 2 2018 MICROSOFT CORPORATION 1251 13 3199 54 51 4442 WA REDMOND 98052 4450 3 2018 INTEL CORPORATION 873 9 1260 19 33 2743 AZ CHANDLER 85248 2133 4 2018 GOOGLE LLC 724 6 1931 17 54 3581 CA MOUNTAIN VIEW 94043 2655 5 2018 ERNST & YOUNG US LLP 716 93 1760 150 54 5596 NJ SECAUCUS 7094 2476 6 2018 APPLE INC 697 13 2386 25 33 4110 CA CUPERTINO 95014 3083 7 2018 FACEBOOK INC 650 5 1406 12 51 5019 CA MENLO PARK 94025 2056 8 2018 DELOITTE CONSULTING LLP 592 295 4193 1281 54 4513 PA PHILADELPHIA 19103 4785 9 2018 TECH MAHINDRA AMERICAS INC 579 201 1781 300 54 2696 NJ SOUTH PLAINFIELD 7080 2360 10 2018 TATA CONSULTANCY SVCS LTD 528 152 8227 1744 54 9806 MD ROCKVILLE 20850 8755Following are the top 10 employers by I.A. (Initial Approvals) shown in the above screenshot:
Employer Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R
1 AMAZON FULFILLMENT SERVICES INC 8 2397 23 1995 45 4392 99.0 97.8 98.5
2 MICROSOFT CORPORATION 12 1256 13 3212 54 4468 99.0 98.3 98.5
3 INTEL CORPORATION 8 875 9 1273 19 2148 99.0 98.5 98.7
4 GOOGLE LLC 14 734 6 1954 17 2688 99.2 99.1 99.2
5 ERNST & YOUNG US LLP 9 717 94 1766 150 2483 88.4 92.2 91.1
6 APPLE INC 11 701 13 2402 26 3103 98.2 98.9 98.8
7 DELOITTE CONSULTING LLP 28 697 323 4735 1461 5432 68.3 76.4 75.3
8 FACEBOOK INC 29 669 7 1821 14 2490 99.0 99.2 99.2
9 TECH MAHINDRA AMERICAS INC 16 587 202 1843 311 2430 74.4 85.6 82.6
10 TATA CONSULTANCY SVCS LTD 21 533 156 8257 1746 8790 77.4 82.5 82.2
As can be seen, the second 10 rows do not contain the FYear, NAICS, Tax.ID, State, City, or ZIP columns. That's because the records have been grouped by just the Employer. However, it contains the additional columns of Records which is simply the number of records that were found and combined for each employer. Also, it contains the new columns I.A.R, C.A.R, and T.A.R. These are simply the Initial Approval Rate, Continuing Approval Rate, and Total Approval Rate as calculated from I.A., I.D., C.A., and C.D.
One item of interest is that the approval rates of the consulting companies (ERNST & YOUNG US LLP, DELOITTE CONSULTING LLP, TECH MAHINDRA AMERICAS INC, and TATA CONSULTANCY SVCS LTD) are all lower than the "regular" 6 tech companies. The lowest Initial Approval Rate for these latter 6 companies is 98.2 percent while the highest Initial Approval Rate for the 4 consulting companies is 88.4 percent. To see if this pattern extends beyond the top 10 companies, change "Sort by" to I.A.R (Initial Approval Rate). In order to focus on the largest users of Initial Approvals, set the "Include in Workers" checkboxes to only include I.A. (Initial Approvals) and set "Minimum Number of Workers" to 150. That causes the following table to be output:
Employer Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R
1 NVIDIA CORPORATION 5 162 0 474 1 162 100.0 99.8 99.8
2 THE LELAND STANFORD JR UNIVERSITY 3 195 1 167 1 195 99.5 99.4 99.5
3 UNIVERSITY OF MICHIGAN 4 180 1 149 2 180 99.4 98.7 99.1
4 ADOBE SYSTEMS INCORPORATED 2 150 1 342 4 150 99.3 98.8 99.0
5 GOOGLE LLC 14 734 6 1954 17 734 99.2 99.1 99.2
6 AMAZON FULFILLMENT SERVICES INC 8 2397 23 1995 45 2397 99.0 97.8 98.5
7 FACEBOOK INC 29 669 7 1821 14 669 99.0 99.2 99.2
8 IBM CORPORATION 15 310 3 821 99 310 99.0 89.2 91.7
9 INTEL CORPORATION 8 875 9 1273 19 875 99.0 98.5 98.7
10 MICROSOFT CORPORATION 12 1256 13 3212 54 1256 99.0 98.3 98.5
11 BLOOMBERG LP 5 225 4 473 20 225 98.3 95.9 96.7
12 APPLE INC 11 701 13 2402 26 701 98.2 98.9 98.8
13 LINKEDIN CORPORATION 6 168 3 561 10 168 98.2 98.2 98.2
14 JPMORGAN CHASE & CO 9 325 8 1881 55 325 97.6 97.2 97.2
15 CUMMINS INC 4 315 11 616 26 315 96.6 96.0 96.2
16 BANK OF AMERICA N A 6 181 7 665 3 181 96.3 99.6 98.8
17 CISCO SYSTEMS INC 9 331 13 1328 28 331 96.2 97.9 97.6
18 AMAZON WEB SERVICES INC 9 249 12 466 29 249 95.4 94.1 94.6
19 UBER TECHNOLOGIES INC 7 152 10 544 10 152 93.8 98.2 97.2
20 AMAZON CORPORATE LLC 10 154 15 826 31 154 91.1 96.4 95.5
21 WAL-MART ASSOCIATES INC 6 342 35 708 35 342 90.7 95.3 93.8
22 HEXAWARE TECHNOLOGIES INC 7 156 20 525 52 156 88.6 91.0 90.4
23 ERNST & YOUNG US LLP 9 717 94 1766 150 717 88.4 92.2 91.1
24 ORACLE AMERICA INC 20 173 26 1489 28 173 86.9 98.2 96.9
25 L&T TECHNOLOGY SERVICES LIMITED 3 253 50 908 102 253 83.5 89.9 88.4
26 DELOITTE & TOUCHE LLP 17 271 58 753 337 271 82.4 69.1 72.2
27 MCKINSEY & CO INC UNITED STATES 1 190 52 176 28 190 78.5 86.3 82.1
28 LARSEN & TOUBRO INFOTECH LIMITED 4 155 43 1287 171 155 78.3 88.3 87.1
29 GOLDMAN SACHS & CO 1 196 55 381 8 196 78.1 97.9 90.2
30 TATA CONSULTANCY SVCS LTD 21 533 156 8257 1746 533 77.4 82.5 82.2
31 WIPRO LIMITED 12 276 83 2885 601 276 76.9 82.8 82.2
32 MPHASIS CORPORATION 9 176 60 918 141 176 74.6 86.7 84.5
33 TECH MAHINDRA AMERICAS INC 16 587 202 1843 311 587 74.4 85.6 82.6
34 ACCENTURE LLP 17 366 162 2694 452 366 69.3 85.6 83.3
35 DELOITTE CONSULTING LLP 28 697 323 4735 1461 697 68.3 76.4 75.3
36 HCL AMERICA INC 11 196 103 2111 510 196 65.6 80.5 79.0
37 COGNIZANT TECH SOLNS US CORP 12 501 791 8762 3553 501 38.8 71.1 68.1
38 CAPGEMINI AMERICA INC 25 273 1068 2681 924 273 20.4 74.4 59.7
All of the employers from HEXAWARE TECHNOLOGIES INC (number 22) on appear to be consulting companies except for ORACLE AMERICA INC and GOLDMAN SACHS & CO. In fact, all but those and MCKINSEY & CO INC UNITED STATES and L&T TECHNOLOGY SERVICES LIMITED appear in the Wikipedia list of IT consulting firms. All of the employers before HEXAWARE TECHNOLOGIES INC appear not to be IT consulting firms. Hence, there does seem to be a lower initial approval rate for consulting firms. The reason for this is not clear from the data. An article titled Avoid These 3 Reasons for H-1B Denials lists those 3 reasons as follows:
In any event, it's also worth noting that the two universities listed, Stanford and Michigan, had initial approval rates of 99.5 and 99.4, respectively. In fact, most of the most recognizable tech firms have initial approval rates of 99 percent and above. Apple and LinkedIn trailed slightly with 98.2 percent and Cisco had 96.2 percent. Hence, the recent increase in denials appear to be chiefly affecting consulting firms. This is not that surprising since such firms tend to hire workers with lower-level skills which may not qualify as unique. Also, there may be more difficulty in showing an employer-employee relationship since the H-1B workers are typically sent to work at client companies.
It should be noted that the two groups of rows contain the same 10 companies though the order of FACEBOOK INC and DELOITTE CONSULTING LLP are swapped. Also, the values of I.A., I.D., C.A., and C.D. in the second 10 rows are all greater than or equal to the corresponding values in the first 10 rows. That is because the values in the second 10 rows equal the corresponding values in the first 10 rows plus the values from other records that matched each Employer name. The largest jump in Workers is for DELOITTE CONSULTING LLP which jumps from 4785 to 5432. A list of all of the matching records for can be seen by entering DELOITTE CONSULTING LLP into "Search Employer", checking the "Output all records" checkbox, and pressing the Submit button. The following data is output:
FYear Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R
1 2018 28 697 323 4735 1461 5432 68.3 76.4 75.3
Employer Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R
1 DELOITTE CONSULTING LLP 28 697 323 4735 1461 5432 68.3 76.4 75.3
FYear Employer I.A. I.D. C.A. C.D. NAICS Tax.ID State City ZIP Workers
1 2018 DELOITTE CONSULTING LLP 592 295 4193 1281 54 4513 PA PHILADELPHIA 19103 4785
2 2018 DELOITTE CONSULTING LLP 99 28 420 145 54 4513 MA BOSTON 2210 519
3 2018 DELOITTE CONSULTING LLP 1 0 1 0 54 4513 PA PHILADELPHIA 19103 2
4 2018 DELOITTE CONSULTING LLP 1 0 0 0 54 1513 PA PHILADELPHIA 19103 1
5 2018 DELOITTE CONSULTING LLP 1 0 1 0 54 4513 TORONTO NA 2
6 2018 DELOITTE CONSULTING LLP 1 0 93 31 54 4513 GA ATLANTA 30309 94
7 2018 DELOITTE CONSULTING LLP 1 0 0 0 51 4513 PA PHILADELPHIA 19103 1
8 2018 DELOITTE CONSULTING LLP 1 0 0 0 54 5413 PA PHILADELPHIA 19103 1
9 2018 DELOITTE CONSULTING LLP 0 0 2 0 54 4513 PA PHILADELPHIA 19103 2
10 2018 DELOITTE CONSULTING LLP 0 0 2 0 54 4513 PA PHILADELPHIA 19103 2
11 2018 DELOITTE CONSULTING LLP 0 0 1 0 54 4451 PA PHILADELPHIA 19103 1
12 2018 DELOITTE CONSULTING LLP 0 0 1 0 54 4512 PA PHILADELPHIA 19103 1
13 2018 DELOITTE CONSULTING LLP 0 0 3 0 54 4513 MA BOSTON 2110 3
14 2018 DELOITTE CONSULTING LLP 0 0 1 0 54 4513 PA NEW CASTLE 16103 1
15 2018 DELOITTE CONSULTING LLP 0 0 1 0 54 4513 PA PHILADELPHIA 19102 1
16 2018 DELOITTE CONSULTING LLP 0 0 1 0 54 4513 PA PHILADELPHIA 19107 1
17 2018 DELOITTE CONSULTING LLP 0 0 2 0 54 4513 PA PHILADELPHIA 19130 2
18 2018 DELOITTE CONSULTING LLP 0 0 1 0 54 4551 PA PHILADELPHIA 19103 1
19 2018 DELOITTE CONSULTING LLP 0 0 1 1 54 4613 PA PHILADELPHIA 19103 1
20 2018 DELOITTE CONSULTING LLP 0 0 1 0 54 4513 PA PHILADELPHIA 19103 1
21 2018 DELOITTE CONSULTING LLP 0 0 2 0 54 5451 PA PHILADELPHIA 19103 2
22 2018 DELOITTE CONSULTING LLP 0 0 5 0 54 4513 PA PHILADELPHIA 19103 5
23 2018 DELOITTE CONSULTING LLP 0 0 1 0 54 4513 PA PHILADELPHIA 19103 1
24 2018 DELOITTE CONSULTING LLP 0 0 1 0 54 817 PA PHILADELPHIA 19103 1
25 2018 DELOITTE CONSULTING LLP 0 0 1 0 54 2549 PA PHILADELPHIA 19103 1
As can be seen, most of the increase in workers was due to 519 workers in Boston and 94 workers in Atlanta. However, also visible are a few apparent typos. Note that all of the Tax.ID values are 4513 except for values of 1513, 5413, 4451, 4512, 4551, 4613, 5451, 817, and 2549. At least the values of 1513, 5413, 4512, and 4613 are likely typos of 4513. Also, the ZIP values of 19102, 19107, and 19130 are likely typos of 19103. The possibility of typos is mentioned in the section "Data Accuracy" at this USCIS document that states "For example, an applicant may mix up numbers in their tax identification number."
As seen in the prior screenshot, the R Shiny app first outputs the totals by year. Hence, the first data output in the screenshot is the following:
FYear Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R 1 2020 55239 122894 18276 303830 21639 426724 87.1 93.4 91.4The R Shiny app can output the data for any span of available years by setting both the Start Year and End Year. For example, setting the Start Year to 2009, the End Year to 2020 and pressing the Submit button causes the initial data output to be the following (this may take 30 seconds or so):
FYear Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R 1 2009 68919 99790 18099 127687 7967 227477 84.6 94.1 89.7 2 2010 55429 84038 7352 108935 5170 192973 92.0 95.5 93.9 3 2011 62874 114863 8583 155186 4548 270049 93.0 97.2 95.4 4 2012 56222 140809 7401 129023 4028 269832 95.0 97.0 95.9 5 2013 56079 130736 9596 164538 4552 295274 93.2 97.3 95.4 6 2014 56595 130616 10645 202190 6120 332806 92.5 97.1 95.2 7 2015 48544 111846 7644 176656 5429 288502 93.6 97.0 95.7 8 2016 53129 105092 11295 252120 11737 357212 90.3 95.6 93.9 9 2017 49786 96166 14518 277229 15338 373395 86.9 94.8 92.6 10 2018 55666 87889 28181 247072 33165 334961 75.7 88.2 84.5 11 2019 59441 132967 35633 256356 33880 389323 78.9 88.3 84.9 12 2020 55239 122894 18276 303830 21639 426724 87.1 93.4 91.4Ensure that "Plot type" is set to "Workers by Status" and press the Plot tab. The following graph will be output:
The sharp decline in 2019 is chiefly due to the fact that this includes just one quarter's worth of data. Other than that, the graph shows that initial approvals have been declining slightly since 2012 and continuing approvals have been generally increasing since then and increased sharply in 2016. One seemingly strange thing is that the graph shows that the number of continuing approvals have increased to be much greater than the initial approvals, especially from 2016 on. It would seem that the main source of applications for a continuing approval would be H-1B workers who received an initial approval 3 years prior (since the initial and continuing H-1B approval is good for 3 years). However, the 252,120 continuing approvals in 2016 is nearly twice as much as the 130,736 initial approvals 3 years earlier. In their article titled "Understanding Our H-1B Employer Data Hub", the USCIS describes Continuing Approval as follows:
H-1B petitions with anything other than "New employment" or "New concurrent employment" selected on Part 2, Question 2 of the Form I-129, whose first decision is an approval. This includes, for example, continuing employment, change of employer, and amended petitions.
Hence, many of the additional H-1B approvals may be for existing H-1B workers who changed their employer or amended their petition.
Changing "Plot type" to "Approval Rates" will output the following graph:
Once again, 2019 includes just one quarter's worth of data so it's value may not be totally indicative of what the final rates for 2019 turn out to be. Still, approval rates have been generally dropping since 2015.
Finally, changing "Plot type" to "Records and Workers" will output the following graph:
Note that the values for Workers is affected by which of the IA, ID, CA, and CD checkboxes are checked under "Include in Workers".
The above graphs can be created for a single employer. For example, set "Search Employer" to "AMAZON FULFILLMENT SERVICES INC" (without the quotes), select "Workers by Status" for "Plot type", and press the Submit button. The following graph will be output:
It would seem that Amazon has been using other names for their company. Change "Search Employer" to AMAZON and press the Submit button. The following graph will be output:
Pressing the Output tab shows that AMAZON matched 61 unique Employer names, some of which are obviously not a part of the company run by Jeff Bezos. Setting the "Minimum Number of Workers" numeric input to 10 causes this list to be cut down to the following list of 25:
Employer Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R
1 AMAZON CORPORATE LLC 113 7577 67 8447 106 16024 99.1 98.8 98.9
2 AMAZON FULFILLMENT SERVICES INC 17 2462 24 2112 46 4574 99.0 97.9 98.5
3 AMAZON WEB SERVICES INC 33 701 21 1080 48 1781 97.1 95.7 96.3
4 AMAZON.COM SERVICES INC 2 607 27 124 27 731 95.7 82.1 93.1
5 AMAZON GLOBAL RESOURCES INC 5 256 0 301 4 557 100.0 98.7 99.3
6 AMAZON.COM.KYDC INC 3 107 0 84 0 191 100.0 100.0 100.0
7 AMAZON.COM.DEDC LLC 12 99 4 113 3 212 96.1 97.4 96.8
8 AMAZON ROBOTICS LLC 5 27 2 68 2 95 93.1 97.1 96.0
9 AMAZON.COM.KYDC LLC 6 26 0 25 1 51 100.0 96.2 98.1
10 AMAZON COM SERVICES INC 20 21 0 9 1 30 100.0 90.0 96.8
11 AMAZON WEB SERVICES LLC 6 21 0 26 0 47 100.0 100.0 100.0
12 AMAZON COM AZDC LLC 6 16 0 11 0 27 100.0 100.0 100.0
13 AMAZON SERVICES LLC 9 13 1 30 2 43 92.9 93.8 93.5
14 AMAZON COM INDC LLC 10 11 1 13 0 24 91.7 100.0 96.0
15 AMAZON.COM KYDC LLC 2 10 0 13 1 23 100.0 92.9 95.8
16 AMAZON MECHANICAL TURK INC 7 8 0 8 0 16 100.0 100.0 100.0
17 AMAZON CAPITAL SERVICES INC 6 4 0 10 1 14 100.0 90.9 93.3
18 AMAZON DIGITAL SERVICES INC 5 4 0 9 2 13 100.0 81.8 86.7
19 AMAZON MEDIA GROUP LLC 6 4 1 13 1 17 80.0 92.9 89.5
20 AMAZON SVCS LLC 3 4 0 14 0 18 100.0 100.0 100.0
21 AMAZON.COM DEDC LLC 2 4 0 20 0 24 100.0 100.0 100.0
22 AMAZON FULFILLMENT SVCS INC 4 2 0 8 0 10 100.0 100.0 100.0
23 AMAZON TECHNOLOGIES INC 3 2 0 10 0 12 100.0 100.0 100.0
24 AMAZON PAYMENTS INC 7 1 0 14 0 15 100.0 100.0 100.0
25 AMAZON MEDIA GROUP 1 0 0 10 0 10 NaN 100.0 100.0
Most, if not all, of these appear to be under the known Amazon. In any event, it appears that the great majority of the data caught by the AMAZON filter are a part of the known Amazon. It is possible to get a graph showing the Workers for each of these different Employer names by selecting "Workers by Employer Name" as the "Plot type" and pressing on the Plot tab. The following graph is output:
As can be seen, the main name that Amazon used in its Form I-129 through 2017 was AMAZON CORPORATE LLC but, in 2018, this name was used far less and AMAZON FULFILLMENT SERVICES INC became the most used name. The third most used name in 2018 was AMAZON WEB SERVICES INC.
Following is a prior list of the employers sorted by I.A. (Initial Approvals):
FYear Employer I.A. I.D. C.A. C.D. NAICS Tax.ID State City ZIP Workers 1 2018 AMAZON FULFILLMENT SERVICES INC 2392 22 1980 45 45 4687 WA SEATTLE 98121 4372 2 2018 MICROSOFT CORPORATION 1251 13 3199 54 51 4442 WA REDMOND 98052 4450 3 2018 INTEL CORPORATION 873 9 1260 19 33 2743 AZ CHANDLER 85248 2133 4 2018 GOOGLE LLC 724 6 1931 17 54 3581 CA MOUNTAIN VIEW 94043 2655 5 2018 ERNST & YOUNG US LLP 716 93 1760 150 54 5596 NJ SECAUCUS 7094 2476 6 2018 APPLE INC 697 13 2386 25 33 4110 CA CUPERTINO 95014 3083 7 2018 FACEBOOK INC 650 5 1406 12 51 5019 CA MENLO PARK 94025 2056 8 2018 DELOITTE CONSULTING LLP 592 295 4193 1281 54 4513 PA PHILADELPHIA 19103 4785 9 2018 TECH MAHINDRA AMERICAS INC 579 201 1781 300 54 2696 NJ SOUTH PLAINFIELD 7080 2360 10 2018 TATA CONSULTANCY SVCS LTD 528 152 8227 1744 54 9806 MD ROCKVILLE 20850 8755In the prior graph titled "H1B Initial and Continuing, Approvals and Denials, Employer contains AMAZON", the initial and continuing approvals seem to be increasing together. However, this was observed to not be the case for many employers. For this reason, it's useful to ensure that only the I.A. and I.C. checkboxes are set in the "Include in Workers" checkboxes and then to set "Sort by" to Workers. Following are the first 20 totals by Employer in 2018:
Employer Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R
1 COGNIZANT TECH SOLNS US CORP 12 501 791 8762 3553 9263 38.8 71.1 68.1
2 TATA CONSULTANCY SVCS LTD 21 533 156 8257 1746 8790 77.4 82.5 82.2
3 INFOSYS LTD 5 69 80 5902 2045 5971 46.3 74.3 73.8
4 DELOITTE CONSULTING LLP 28 697 323 4735 1461 5432 68.3 76.4 75.3
5 MICROSOFT CORPORATION 12 1256 13 3212 54 4468 99.0 98.3 98.5
6 AMAZON FULFILLMENT SERVICES INC 8 2397 23 1995 45 4392 99.0 97.8 98.5
7 WIPRO LIMITED 12 276 83 2885 601 3161 76.9 82.8 82.2
8 APPLE INC 11 701 13 2402 26 3103 98.2 98.9 98.8
9 ACCENTURE LLP 17 366 162 2694 452 3060 69.3 85.6 83.3
10 CAPGEMINI AMERICA INC 25 273 1068 2681 924 2954 20.4 74.4 59.7
11 GOOGLE LLC 14 734 6 1954 17 2688 99.2 99.1 99.2
12 FACEBOOK INC 29 669 7 1821 14 2490 99.0 99.2 99.2
13 ERNST & YOUNG US LLP 9 717 94 1766 150 2483 88.4 92.2 91.1
14 TECH MAHINDRA AMERICAS INC 16 587 202 1843 311 2430 74.4 85.6 82.6
15 HCL AMERICA INC 11 196 103 2111 510 2307 65.6 80.5 79.0
16 JPMORGAN CHASE & CO 9 325 8 1881 55 2206 97.6 97.2 97.2
17 INTEL CORPORATION 8 875 9 1273 19 2148 99.0 98.5 98.7
18 ORACLE AMERICA INC 20 173 26 1489 28 1662 86.9 98.2 96.9
19 CISCO SYSTEMS INC 9 331 13 1328 28 1659 96.2 97.9 97.6
20 IBM INDIA PRIVATE LIMITED 11 62 60 1562 289 1624 50.8 84.4 82.3
As can be seen, this is a very different list than the list sorted by I.A. (Initial Approvals). In the prior list, only 4 of the top 10 employers were consulting firms and those were in positions 5 and 8 through 10. In this list, 7 of the top 10 employers are consulting firms and those are in positions 1 through 4, 7, 9, and 10. Looking at the approvals in the list, it appears that the consulting firms tend to have many more continuing approvals than initial approvals in 2018, sometimes as much as 10 times as many. Following are graphs of the approvals over time of the first 12 employers in the list above. The graphs just include the complete fiscal years (through 2018) and the graphs were made less wide by simply making the browser window that holds them less wide.
As can be seen, the initial approvals of all of the consulting firms are decreasing toward zero and the continuing approvals were going up through 2017 but decreased in 2018 (except ACCENTURE which started decreasing in 2017). The initial approvals of the non-consulting firms are generally moving up though 3 in 5 went down in 2018. Their continuing approvals are also generally moving up and 3 out of 5 (APPLE, GOOGLE, and FACEBOOK) increased sharply in 2018. It's not clear what is going on here but it is possible that the non-consulting firms are hiring away many of the H-1B workers from the consulting firms.
The third and fourth tables in this document show that INTEL CORPORATION had the third most Initial Approvals in 2018. In trying to find all of the records associated with that company, simply searching for INTEL matches numerous unassociated records. Some experimentation showed that searching for "^INTEL " (with the trailing blank but without the quotes) matched the following employer names:
Employer Records I.A. I.D. C.A. C.D. Workers I.A.R C.A.R T.A.R
1 INTEL CORPORATION 99 7920 38 9963 189 17883 99.5 98.1 98.7
2 INTEL MOBILE COMMUNICATIONS 13 154 3 412 4 566 98.1 99.0 98.8
3 INTEL CORP 39 89 0 65 2 154 100.0 97.0 98.7
4 INTEL MASSACHUSETTS INC 3 85 2 42 0 127 97.7 100.0 98.4
5 INTEL AMERICAS 8 32 1 67 2 99 97.0 97.1 97.1
6 INTEL AMERICAS INC 1 26 2 40 1 66 92.9 97.6 95.7
7 INTEL FEDERAL LLC 6 13 0 13 1 26 100.0 92.9 96.3
8 INTEL FEDERAL 1 3 0 12 0 15 100.0 100.0 100.0
9 INTEL FEDERAL INC 1 3 0 7 1 10 100.0 87.5 90.9
10 INTEL GE CARE INNOVATIONS LLC 3 3 1 2 0 5 75.0 100.0 83.3
11 INTEL MASSACHUSETTS 5 2 0 7 0 9 100.0 100.0 100.0
12 INTEL TUTORS INC 1 1 0 0 0 1 100.0 NaN 100.0
13 INTEL CORPOTATION 1 0 0 1 0 1 NaN 100.0 100.0
14 INTEL MONTESSORI ACADEMY OF MINNES 1 0 0 1 0 1 NaN 100.0 100.0
15 INTEL PROFESSIONAL REALTY INC 1 0 0 1 0 1 NaN 100.0 100.0
16 INTEL RECRUITING NETWORK DBA REHAB 1 0 0 3 0 3 NaN 100.0 100.0
17 INTEL SECURITY PUBLIC SECTOR LLC 1 0 0 1 0 1 NaN 100.0 100.0
It appears that the employer names through INTEL MASSACHUSETTS are associated with the company so this search seems to be a good estimate of total H-1B approved for Intel Corporation. As can be seen in the (Wikipedia page for Regular expression, the leading carat (^) is a metacharacter that matches the starting position within the string. In fact, the searches in this application use regex (regular expressions). Following are the graphs output for this search and Plot types of "Worker by Status" and "Workers by Employer Name":
Another employer that required regex metacharacters was Wal-Mart. It appears that both WALMART and WAL-MART have been used in the applications. In addition, it appears that WAL-MART was misspelled as WALT-MART in 2017. All of these spellings can be caught by the regex string WAL[A-Z]?[-]?MART or by the slightly simpler WAL[A-Z-]*MART. Following are the graphs output for this search and Plot types of "Worker by Status" and "Workers by Employer Name":
As can be seen in the second graph, WAL-MART ASSOCIATES INC seems to have simply been mistyped with an extra T in 2017.
This application is under continuing development. If anyone should run into any issues or have any suggestions for additional features, feel free to let me know via the Contact box at the bottom of this page.