EB2-3 Predictions (Rather Calculations) - 2014

[sbhagwat2000]

I'm not aware of anything special about FY2011 off hand. FY2010 was a little below average and the combination of the two was about average.

In FY2012, EB2-WW were retrogressed for 3 months, so in FY2013, potentially EB2-WW approvals can be a normal year's worth plus the additional 6-7k who could not be approved in FY2013 due to retrogression.

For that reason, it may be difficult to get any spillover from EB2-WW this year and in a worse case scenario EB2-WW would actually use any spillover from EB1.

Even in a zero spillover scenario, which at this time I don't actually think will happen, the Cut Off Dates should still move past Feb 2005 at some point. The trick would to get approved while they remain so.

Spec,

Thanks for your reply. We will have to see EB1 usage this year to see how much fall down it gives. I had one more question which is bothering me. The Demand data pre 2007 which fell from 1350 to 1100. does that contain the iterfiling requests made between may and oct when the dates were U. Or those will be added now

[PD2008AUG25]

For that reason, it may be difficult to get any spillover from EB2-WW this year and in a worse case scenario EB2-WW would actually use any spillover from EB1.

Wouldn't any spillover from EB1 go to most retrogressed country, India? I thought spillover is only based on priority date.

[sbhagwat2000]

I was reading up on the visa statistics for the past years and see that a couple of years - 2009 and 2010 EB1 actually used 41000 visas. Thats a bummer. I guess 2013 is building out to be a perfect storm. EB1 demand is high so will most likely consume its quota. EB5 - theres going to be a COD. EB2 WW due to fall through from last year may consume or exceeed its quota. So with this are there any other places from where visas can be spilled overto EB2 I? I hope I am wrong somewhere here but looks like this year theres going to be no spillover. With no SO we are left to the unknowns of demand pre 2004 and so can easily see dates not moving at all. But whats more concerning is that this looks like a long term trend and hence EB2 I will get only 2800 visas for the next few years. Gurus - please correct me if I am wrong with my analysis.

[bvsamrat]

Can you also check how many Eb2-ROW visas were given out on 2009/10/11?
This might give us a clue?

I was reading up on the visa statistics for the past years and see that a couple of years - 2009 and 2010 EB1 actually used 41000 visas. Thats a bummer. I guess 2013 is building out to be a perfect storm. EB1 demand is high so will most likely consume its quota. EB5 - theres going to be a COD. EB2 WW due to fall through from last year may consume or exceeed its quota. So with this are there any other places from where visas can be spilled overto EB2 I? I hope I am wrong somewhere here but looks like this year theres going to be no spillover. With no SO we are left to the unknowns of demand pre 2004 and so can easily see dates not moving at all. But whats more concerning is that this looks like a long term trend and hence EB2 I will get only 2800 visas for the next few years. Gurus - please correct me if I am wrong with my analysis.

[geevikram]

I was reading up on the visa statistics for the past years and see that a couple of years - 2009 and 2010 EB1 actually used 41000 visas. Thats a bummer. I guess 2013 is building out to be a perfect storm. EB1 demand is high so will most likely consume its quota. EB5 - theres going to be a COD. EB2 WW due to fall through from last year may consume or exceeed its quota. So with this are there any other places from where visas can be spilled overto EB2 I? I hope I am wrong somewhere here but looks like this year theres going to be no spillover. With no SO we are left to the unknowns of demand pre 2004 and so can easily see dates not moving at all. But whats more concerning is that this looks like a long term trend and hence EB2 I will get only 2800 visas for the next few years. Gurus - please correct me if I am wrong with my analysis.

There are too many unknowns. Take it easy. As long as EBWW is current, we really won't know how far the dates will progress.(it may move a lot, it may not move at all) It's anybody's guess.

[vizcard]

I think it's fairly safe to say dates will progress to dec2007. The big variable is porting. Ironically, this time last year,the gurus were predicting dec2007 to mar2008.

[qesehmk]

Agree vizcard. Last year that prediction actually was spot on - except that we assumed FIFO but CO approved 1/3rd of 2008 EB2 cases ahead of 2007 cases.

This year - IMHO - Dec 2007 is absolute safe bet i.e 100% possibility. I would place median probability at Aug 2008.

I think it's fairly safe to say dates will progress to dec2007. The big variable is porting. Ironically, this time last year,the gurus were predicting dec2007 to mar2008.

[vizcard]

Agree vizcard. Last year that prediction actually was spot on - except that we assumed FIFO but CO approved 1/3rd of 2008 EB2 cases ahead of 2007 cases.

This year - IMHO - Dec 2007 is absolute safe bet i.e 100% possibility. I would place median probability at Aug 2008.

Cant blame CO for that Anyway, some folks got lucky, others got shafted but it'll correct itself with time. I'm not so sure about median being Aug 2008. I'd offer that, that is a best case scenario. Personally, Id love that to be the median scenario because of my PD but realistically I'm not sure.

[sbhagwat2000]

Cant blame CO for that Anyway, some folks got lucky, others got shafted but it'll correct itself with time. I'm not so sure about median being Aug 2008. I'd offer that, that is a best case scenario. Personally, Id love that to be the median scenario because of my PD but realistically I'm not sure.

So the years 2009 and 2010 when EB1 used 40000 visas EB2 ROW used 27000 and 30000 respectively and hence we got extra visas. in 2011 when EB2 ROW used more that 34,400 EB1 used 27000. So we were lucky last few years when one used more visas other was using less. Thats 100% not going to happen this year. Best case is EB1 is going to use 40000 and EB2 ROW 34400. So I really dont see any SO this year. The data is correct I may be wrong in analysis and there may be some category I am not aware of but we can say good bye to SOFAD this year from EB1 and EB2 ROW. So in that case I think dates may not move at all or based on current DD move to 2006 start.

[qesehmk]

viz - I would absolutely blame CO for that. Simply because it is his job to ensure that people from same chargeability generally get GC in FIFO manner based on PD.

He should've easily foreseen the 2007 august cliff for EB2IC and moved dates quite early to ensure he had sufficient applications to approve if and when numbers were available. Instead he moved the dates quite haphazard.

As per Aug 2008 being a median or not - I would totally respect your opinion since we all are guessing (somewhat intelligently!). My optimism stems from the continued malaise in economy which means much lower levels of EB2ROW and possibly even EB1. This has been quite true for last 4 years and generally EB2IC have benefitted. I do agree that as economy comes back - EB2IC timelines are going to be crushed significantly.

Cant blame CO for that Anyway, some folks got lucky, others got shafted but it'll correct itself with time. I'm not so sure about median being Aug 2008. I'd offer that, that is a best case scenario. Personally, Id love that to be the median scenario because of my PD but realistically I'm not sure.

[Spectator]

Based mainly on Trackitt numbers and some educated guesses, here is what I think happened in October.

Generally, a few late updates will increase the figure slightly.

EB1 ---- 2,300
EB2-WW - 3,900
EB2-I ---- 800
EB2-C ---- 250
EB3 ---- 3,000
EB4 ------ 850
EB5 ---- 1,000

Total - 12,100

Max Allowed in month (9%) - 12,600

The figures appear to be at least sensible, but I make no claims to their accuracy.

[openaccount]

Based mainly on Trackitt numbers and some educated guesses, here is what I think happened in October.

Generally, a few late updates will increase the figure slightly.

EB1 ---- 2,300
EB2-WW - 3,900
EB2-I ---- 800
EB2-C ---- 250
EB3 ---- 3,000
EB4 ------ 850
EB5 ---- 1,000

Total - 12,100

Max Allowed in month (9%) - 12,600

The figures appear to be at least sensible, but I make no claims to their accuracy.

Spec,

Nice to see some numbers after along time, especially with 485-Inventory delay.

If EB2I used 800 in October itself, then for December also EB2I is going to stay at 2004. Monthly allocation is 250 and in one month EB2I used 3 months numbers.

[Spectator]

Thanks spec. I am very curious as to your methods for arriving at these numbers, but it is interesting.

sportsfan,

You're welcome.

How do you deduce EB2-I used 800 numbers? I thought they were going by a monthly allocation. Does this mean we can expect a fair jump in the December bulletin (that seems contrary to the last word from CO).

I would caution at taking the figures to literally. They are only meant to be indicative.

For EB2-I, I am looking at the historical % between Trackitt approvals and actual numbers. I have to look at both EB2 and EB3 because most porters already had an I-485 pending in EB3. Fortunately they are quite similar. This gives a spread of 750 - 825.

I only look at Primary approvals in Trackitt and calculate from that baseline.

In October, to date, there are 37 EB2-I Trackitt approvals (corrected for those that still show in EB3 etc).

The breakdown by PD for the EB2-I approvals is:

PD --- No.
2001 --- 2
2002 --- 2
2003 -- 19
2004 -- 14

Total - 37

If there were only 250 approvals, then Trackitt would have to represent about 15% of all approved EB2-I cases, which is far higher than the historical number over several years. It is possible the % has risen slightly and I will amend the numbers if better info becomes available.

I don't think there is any way it is as low as 250. Given the numbers, there may be no further movement in the December VB either. That said, CO works in mysterious ways!

EB2-WW is mostly backlog, so if we subtract that from the equation, then both EB1 and EB2-WW point to a fairly large spillover later this year. What's your take on that?

That's true, but it is a constrained backlog to January 1, 2012.

November will also release more substantial EB2-WW backlog.

I think it is too early to say anything about EB2-WW, given the unusual circumstances.

I also think it is too early to talk about EB1. The approvals can be quite "lumpy". The numbers are also dominated by Indian EB1C approvals, so most EB1 approvals are quite poorly represented, especially EB1-WW in general.

The October EB1 approvals are slightly higher than the same period last year.

EB5 number is absolutely crazy. How did you arrive at that number? Looks like there is not going to be any spillover whatsoever from EB5 this year.

CO said that 10% of EB5 would be used in October. That is 940, so I have rounded up slightly.

EB4 is a pure guess based on maximum usage. I need to see what happened in FY2012. FY2011 was low, but it might have been a blip.

You can find more detail here.

Hope that helps.

[openaccount]

from trackitt, murthy's article :http://www.murthy.com/2012/11/02/pri...ions-for-fy13/

Excess Visa Numbers Estimated for EB2 India and China
Overall, the expectations are that EB2 India and China will benefit from some excess visa numbers shifting from EB1 and EB2 ROW. However, the volume of such excess is expected to be even less in FY13 than it was in FY12. In FY12, EB2 India and China each received approximately 19,000 visa numbers. The vast majority of these were due to shifting excess numbers into EB2 India and China. At this time, the DOS estimates that a total of 13,000 visa numbers will be available to EB2 India and China in FY13. This is subject to change - either up or down - depending upon the usage in the other categories, as explained here.

In FY12, EB2 India and China each received approximately 19,000 visa numbers.
38k total to EB2IC in FY2012 WOW really

At this time, the DOS estimates that a total of 13,000 visa numbers will be available to EB2 India and China in FY13.
which is 10k to EB2I that should take EB2I to Jan2008 assuming porting is <=3k

[qesehmk]

Friends - sorry to steal attention from this thread. But can you please suggest charities for November donations using this site's ad revenues please? We will have about $100 to donate.

please reply HERE http://www.qesehmk.org/forums/showth...-We-Made/page3

[GhostWriter]

Thanks for sharing openaccount. Two more interesting things from the article

1. The estimate given as a best case scenario is that, by the end of FY13, EB2 India's cutoff date will advance to a new cutoff date, set at some point in early 2008.

2. Each month, the DOS reports receiving thousands of these requests connected to upgraded cases. Since EB3 is oversubscribed, regardless of country of chargeability, many of these cases are filed by those in EB3 ROW. .... The increase in cases shifting from EB3 to EB2 ROW increases demand in that category. The result is fewer excess numbers that can be shifted out of EB2 ROW to EB2 India and China.

from trackitt, murthy's article :http://www.murthy.com/2012/11/02/pri...ions-for-fy13/

[Spectator]

from trackitt, murthy's article :http://www.murthy.com/2012/11/02/pri...ions-for-fy13/

Excess Visa Numbers Estimated for EB2 India and China
Overall, the expectations are that EB2 India and China will benefit from some excess visa numbers shifting from EB1 and EB2 ROW. However, the volume of such excess is expected to be even less in FY13 than it was in FY12. In FY12, EB2 India and China each received approximately 19,000 visa numbers. The vast majority of these were due to shifting excess numbers into EB2 India and China. At this time, the DOS estimates that a total of 13,000 visa numbers will be available to EB2 India and China in FY13. This is subject to change - either up or down - depending upon the usage in the other categories, as explained here.

In FY12, EB2 India and China each received approximately 19,000 visa numbers.
38k total to EB2IC in FY2012 WOW really

At this time, the DOS estimates that a total of 13,000 visa numbers will be available to EB2 India and China in FY13.
which is 10k to EB2I that should take EB2I to Jan2008 assuming porting is <=3k

openaccount,

Thanks for that. Unfortunately , some of the language is quite ambiguous. I very much doubt that EB2-C received 19k last year. I think they mean EB2-IC received 19k between them. If so, that is quite a bit lower than I was expecting. 19k + 5.8k would have made sense.

At a 13k SOFAD level, EB2-C wouldn't use any spillover, so EB2-I would have 10.2k available. That is pretty much what I have been using to date.

That might leave EB2-I in Q4 2007, possibly very early 2008, depending on porting levels.

If the number is talking about 13k spillover then dates could move to April 2008 for EB2-I. At the moment, 13k SO seems a little optimistic.

[justvisiting]

openaccount,

Thanks for that. Unfortunately , some of the language is quite ambiguous. I very much doubt that EB2-C received 19k last year. I think they mean EB2-IC received 19k between them. If so, that is quite a bit lower than I was expecting.

At a 13k SOFAD level, EB2-C wouldn't use any spillover, so EB2-I would have 10.2k available. That is pretty much what I have been using to date.

That might leave EB2-I in Q4 2007, possibly very early 2008, depending on porting levels.

Porting has become CO's [and everyone trying to predict PD movement]'s greatest headache. It seems pretty clear by now he has no idea how much is out there. This is surprising to me, and it all goes back to USCIS not compiling ANY statistics on I-140 approvals.

There haven't been any Inventory updates either...

[bvsamrat]

I have seen here and other forums that many enquiring with PDs of 2004/2005 are majority porters. But very few porters with PDs of 2006/7/8. If at all , will all these cases be added to demand at any time in future?

Porting has become CO's [and everyone trying to predict PD movement]'s greatest headache. It seems pretty clear by now he has no idea how much is out there. This is surprising to me, and it all goes back to USCIS not compiling ANY statistics on I-140 approvals.

There haven't been any Inventory updates either...

[GhostWriter]

I think the focus has been off the correct type of porting. If the Murthy article is accurate then bulk of the porting is EB3-ROW porting. If we take what they call "thousands of upgrade applications per month" as just 1K per month that is 12K porting per year. If EB3-I porting is 3K per year that still leaves room for 9K EB3-ROW porting. This actually also explains what we were discussing couple of pages ago i.e. why does low PERM applications for ROW not reconcile with high number of EB2-ROW I-485s. I think high EB3-ROW porting explains that anomaly. PD for EB3-ROW has been between 2004-2006 last couple of years. Post 2007 EB3-ROWers (without EADs) would have been porting over this time and since EB2-ROW has been mostly current they would never have shown up in the demand data.

Since details for porting are not available this is just a theory and could be wrong but i do think that low ROW PERM numbers but high EB2-ROW usage does make it seem possible. So EB2-I could have been losing much more potential spillover to EB3-ROW (and not to EB3-I).

I have seen here and other forums that many enquiring with PDs of 2004/2005 are majority porters. But very few porters with PDs of 2006/7/8. If at all , will all these cases be added to demand at any time in future?

[Spectator]

I think the focus has been off the correct type of porting. If the Murthy article is accurate then bulk of the porting is EB3-ROW porting. If we take what they call "thousands of upgrade applications per month" as just 1K per month that is 12K porting per year. If EB3-I porting is 3K per year that still leaves room for 9K EB3-ROW porting. This actually also explains what we were discussing couple of pages ago i.e. why does low PERM applications for ROW not reconcile with high number of EB2-ROW I-485s. I think high EB3-ROW porting explains that anomaly. PD for EB3-ROW has been between 2004-2006 last couple of years. Post 2007 EB3-ROWers (without EADs) would have been porting over this time and since EB2-ROW has been mostly current they would never have shown up in the demand data.

Since details for porting are not available this is just a theory and could be wrong but i do think that low ROW PERM numbers but high EB2-ROW usage does make it seem possible. So EB2-I could have been losing much more potential spillover to EB3-ROW (and not to EB3-I).

GhostWriter,

That is not how I read the article.

I believe the bulk of porting comes from India, at a level much higher than you have indicated, for several reasons. I won't discuss that.

There certainly is ROW-M-P porting, but I think it is (currently) at a lower level.

Remember that CO really doesn't have any idea of the numbers.

For EB2-ROW / EB2-NIW-ROW (doesn't include M or P) adjusted Trackitt approvals to date, here is the PD year breakdown:

2007 -- 3
2008 -- 1
2009 -- 4
2010 -- 6
2011 - 61
2012 - 22

Total - 97

Some of the 2007 numbers (2) relate to Indians with Cross Chargeability to their spouse.

A further 7 of the cases 2007 through 2010 appear to be porting. Other cases appear to be caused by the long time it took to get PERM and I-140 approval. For some cases,it is not possible to tell due to lack of information.

The 2011 and 2012 PD cases are quite expected.

Hope the figures help.

[GhostWriter]

GhostWriter,

There certainly is ROW-M-P porting, but I think it is (currently) at a lower level.

Spec, below are the EB-WW (EB-ROW-MP) PERM numbers, expected I-485s using these numbers and actual EB2-WW approvals. The difference should be an indicator of porting. While all the PERMs filed in one year don't get GC issued in the same year, but since EB2-WW is current it should be within two years from PD in most cases. The average over years should even out the timing issue.

I have assumed 70% of PERMs filed in EB-WW to be for EB2, 90% approval for I-140 (vs. 80% you assume for EB2-IC analysis) and dependent factor of 2. Also assumed zero demand destruction. This gives an overall factor of 1.26. Even with this factor there is a high gap in expected and actual approvals in EB2-WW for 2011 and 2012. So there is a possibility that there is high EB3-WW to EB2-WW porting at least since last two years.

(Rounded to nearest 1000)
Year---PERM-------Exp.485-----Approvals----Porting
2009---23K---------29K---------33K---------4K (~3.7K)
2010---19K---------24K---------27K---------4K (~3.8K)
2011---18K---------23K---------35K---------12K (~11.6K)
2012---22K---------27K---------34K*---------7K (~7.3K)
Avg.---20K---------26K---------32K---------7K (~6.6K)

* For 2012 approvals number is an estimate assuming no cut-off date were imposed on EB-WW. For 2009-2011 the approvals are actual numbers from the DOS visa reports (link).

The average porting estimate then for EB3-WW is 6.6K over last four years and it is 9.5K over last two years !!! Do you have any other explanation for the gap in the above table. 2008 did have a high number of (38K) PERMs for EB-WW but they would have been approved by the end of 2010 (at the latest).

The 100 sample points from Trackitt that you provided are giving us a range of +/- 3500. So if we take 11 cases (out of 9 certain and 5 suspects) as porting cases in your sample of 100 then it is telling us that actual porting number will be contained in the interval 350 to 7350 in total of 35000 approvals (with a fairly high level of certainity). The 7.3K number for 2012 in the table above does fall in your range estimate from Trackitt.

What would be your estimate of EB3-WW porting ?

[suninphx]

Spec, below are the EB-WW (EB-ROW-MP) PERM numbers, expected I-485s using these numbers and actual EB2-WW approvals. The difference should be an indicator of porting. While all the PERMs filed in one year don't get GC issued in the same year, but since EB2-WW is current it should be within two years from PD in most cases. The average over years should even out the timing issue.

I have assumed 70% of PERMs filed in EB-WW to be for EB2, 90% approval for I-140 (vs. 80% you assume for EB2-IC analysis) and dependent factor of 2. Also assumed zero demand destruction. This gives an overall factor of 1.26. Even with this factor there is a high gap in expected and actual approvals in EB2-WW for 2011 and 2012. So there is a possibility that there is high EB3-WW to EB2-WW porting at least since last two years.

(Rounded to nearest 1000)
Year---PERM-------Exp.485-----Approvals----Porting
2009---23K---------29K---------33K---------4K (~3.7K)
2010---19K---------24K---------27K---------4K (~3.8K)
2011---18K---------23K---------35K---------12K (~11.6K)
2012---22K---------27K---------34K---------7K (~7.3K)
Avg.---20K---------26K---------32K---------7K (~6.6K)

The average porting estimate then for EB3-WW is 6.6K over last four years and it is 9.5K over last two years !!! Do you have any other explanation for the gap in the above table. 2008 did have a high number of (38K) PERMs for EB-WW but they would have been approved by the end of 2010 (at the latest).

The 100 sample points from Trackitt that you provided are giving us a range of +/- 3500. So if we take 11 cases (out of 9 certain and 5 suspects) as porting cases in your sample of 100 then it is telling us that actual porting number will be contained in the interval 350 to 7350 in total of 35000 approvals (with a fairly high level of certainity). The 7.3K number for 2012 in the table above does fall in your range estimate from Trackitt.

What would be your estimate of EB3-WW porting ?

EB2WW got 34K visas in 2012? Really? The SOFAD number for EB2IC do not match with above numbers IMO.

[kd2008]

Spec, below are the EB-WW (EB-ROW-MP) PERM numbers, expected I-485s using these numbers and actual EB2-WW approvals. The difference should be an indicator of porting. While all the PERMs filed in one year don't get GC issued in the same year, but since EB2-WW is current it should be within two years from PD in most cases. The average over years should even out the timing issue.

I have assumed 70% of PERMs filed in EB-WW to be for EB2, 90% approval for I-140 (vs. 80% you assume for EB2-IC analysis) and dependent factor of 2. Also assumed zero demand destruction. This gives an overall factor of 1.26. Even with this factor there is a high gap in expected and actual approvals in EB2-WW for 2011 and 2012. So there is a possibility that there is high EB3-WW to EB2-WW porting at least since last two years.

(Rounded to nearest 1000)
Year---PERM-------Exp.485-----Approvals----Porting
2009---23K---------29K---------33K---------4K (~3.7K)
2010---19K---------24K---------27K---------4K (~3.8K)
2011---18K---------23K---------35K---------12K (~11.6K)
2012---22K---------27K---------34K---------7K (~7.3K)
Avg.---20K---------26K---------32K---------7K (~6.6K)

The average porting estimate then for EB3-WW is 6.6K over last four years and it is 9.5K over last two years !!! Do you have any other explanation for the gap in the above table. 2008 did have a high number of (38K) PERMs for EB-WW but they would have been approved by the end of 2010 (at the latest).

The 100 sample points from Trackitt that you provided are giving us a range of +/- 3500. So if we take 11 cases (out of 9 certain and 5 suspects) as porting cases in your sample of 100 then it is telling us that actual porting number will be contained in the interval 350 to 7350 in total of 35000 approvals (with a fairly high level of certainity). The 7.3K number for 2012 in the table above does fall in your range estimate from Trackitt.

What would be your estimate of EB3-WW porting ?

From initial reactions, your theory is wrong. For each and every porting case a new PERM is needed. Since, the PERM #s are consistently low, your theory does not hold. Also, EB3-ROW should show proportional lowering of demand - which it has not. One reason might be NIWs and an Indian interfiling for a different country of chargeability due to marriage to a non-Indian - which could result in larger demand for ROWs, and would bake in some factor of demand destruction for India.

[GhostWriter]

EB2WW got 34K visas in 2012? Really? The SOFAD number for EB2IC do not match with above numbers IMO.

Suninphx, For 2012 EB2WW got about 27K approvals in 9 months (as per Spec's estimate) before a cut off date was imposed. Prorated that would have easily crossed 34K. Alternatively you can add the 4K from the demand data and add another 3K for the people who were not able to file in last three months. Either way if a cut off date was not imposed EB2WW would have used entire 34K allotment. I will edit the post to note this point.
For 2011 the approvals are actual and can be found from the visa report (link). Total of 67K EB2 approvals of which 32K went to Eb2-IC and 35K to EB2-ROW. EB2-ROW actually used a little spillover in 2011. I haven't even added the CP cases in these approvals but they won't be a lot.

From initial reactions, your theory is wrong. For each and every porting case a new PERM is needed. Since, the PERM #s are consistently low, your theory does not hold. Also, EB3-ROW should show proportional lowering of demand - which it has not. One reason might be NIWs and an Indian interfiling for a different country of chargeability due to marriage to a non-Indian - which could result in larger demand for ROWs, and would bake in some factor of demand destruction for India.

kd, i am only trying to understand the big gap between PERM and I485 for ROW. Can you explain the 19K gap in 2011 and 2012. I agree with you on PERM but the porting cases will not necessarily reduce EB3-WW demand. If these cases are post 2007 then they would not have filed I-485 yet in EB3. Is your alternate theory an increase in inter-country marriages in last two years ?