Friday, 12 April 2013

CIA World Factbook Updates Fertility Rate Figures

The CIA World Factbook has updated its pages - unnoticed by everyone - to give their 2013 estimates of fertility rate in each country.

The total fertility rate in the area the Factbook calls the West Bank has dropped to 2.91 children born per woman; last year it was 2.98. (Click here, then click on the "People and Society" section to see this year's figure; see here for the historical figures.)

They have also recorded a smaller drop in the fertility rate in Israel: from 2.67 last year to 2.65 this year.

In Gaza, too, it has dropped - from 4.57 to 4.41.

The net result is a narrowing of the fertility gap between Israel and the Palestinian Territories, in particular between Israel and the areas of Judea and Samaria which are now just 0.26 children per mother apart.

I don't know how the CIA gather their figures, or why they differ significantly from the Israeli Central Bureau of Statistics' figures (which, in 2012, gave a total fertility of 3.00 in Israel), but I assume they're based on on real-life data and not just on mathematical models. As I have noted previously, it's unclear if the World Factbook includes the Jewish residents of Judea and Samaria (who have a very high fertility rate) in the West Bank figures or the Israel figures.

We will find out if their measurement of a drop in fertility in Israel is ratified by the Central Bureau of Statistics - and whether that's specific to Jews, Arabs or both - when they release their annual Statistical Abstract, in September.

Thursday, 14 February 2013

Women of the Wall

This article was originally published in D'Varsity (Cambridge Jewish student newsletter) with the title "What Really Grinds my Girsah"

My Girsah has been Ground this week by what can only be described as a fundamental abuse of human rights and capitulation to the tyranny of the majority. I refer of course to the Women of the Wall, who will attempt, once again, to conduct prayers at the kotel hama’aravi on rosh hodesh. I was calmly perusing the internet with a distinctly Unground Girsah when I was suddenly struck by a desire to know how “Women of the Wall” is written in Hebrew. Apparently it’s “nashot hakotel” [sic]. Their justification?

“In Hebrew, the word for women is nashim. Since -im is generally a masculine plural ending and -ot is generally the feminine plural ending, nashim is an exception to the linguistic rule. We chose to use nashot, similar to the way some American feminists have chosen to use womyn for woman and wimmin for women. It is a pro-female assertion that seeks to remove the linguistic dependency of the word woman or women on the word man or men, since unfortunately these female words have largely and historically been characterized as a derivative of the male, a statement which has social implications.
“For more information, please contact us.”

Laughably ridiculous, but at least there is method in their madness; in the words of Forrest Gump, “Stupid is as stupid does”. Anyway, I clicked on the link and sent them a polite and judicious email:

Dear Women of the Wall,

I am perhaps not the first to call you out on this, but I believe that the grammar in your name may be wrong. I don't object to your use of the term "nashot" instead of "nashim". However, if you are going to use the term "nashot", please be aware that you are using it in construct form (semichut), as you are saying "Women OF the Wall", and therefore it should be "neshot" with a sheva under the nun. Please correct me if I am mistaken.

Kol Tuv,

R

After a full day with no reply, I petitioned them again, this time with backup:

Dear Women of the Wall,

Further to my email of the 4th of February, I have found an explicit quotation from the Academy of the Hebrew Language outlining the plural construct form of "women" in Hebrew; please see the following link:
http://hebrew-academy.huji.ac.il/hahlatot/GrammarDecisions/TheNounRhymeAndshape/Pages/Ch2D111.aspx
"The construct form of nashim can be neshei and also neshot."

Best wishes,

R

I am still yet to receive a reply. Like talking to a brick wall. In my opinion – and I’m not known for having particularly strong opinions – it’s outrageous to expect public support when you can’t even adhere to the basic rules of Hebrew grammar. I will keep you posted on developments.

Friday, 18 January 2013

Final Israeli Election Predictions

This article was originally published in D'Varsity (Cambridge Jewish student newsletter) with the title "What Really Grinds my Girsah"

Since Ben-Gurion famously asked his entire cabinet to Hebraize their surnames, nobody* has been elected prime minister of Israel without having at least a Hebrew-sounding surname. Some of my favourite names include Eshkol, Meir, Shamir, and Peres. Given this well-attested premise, let us evaluate some candidates’ chances of winning next week’s election.

Netanyahu, Binyamin (Likud): 63%. Named after the Israeli seaside resort, Netanya, the electorate are likely to make his party the largest because of his beautifully Hebraized surname. Notice the “yahw” at the end: yes, that’s God’s name.

Yachimovich, Shelly (Ha’Avodah): 1%. Poor show on both names. Those who claim that ‘Shelly’ is Hebrew for ‘mine’ show their immeasurable ignorance (it derives originally from ‘Rachel’, but that’s pretty indirect). Her only chance lies in promising to Hebraize her name upon entering office, like President Ephraim Katzir.

Lapid, Yair (Yesh Atid): 15%. A surprisingly popular new party, though we are not surprised, given his excellent surname meaning ‘torch’ (Zech. 12:6) matching ‘Yair’.

Gal-On, Zahava (Meretz): 9%. Representing far-left Meretz, she is not the pundits’ favourite, but we know better. Original name was Schnipitzky: what an improvement.

Bennett, Naftali (HaBayit HaYehudi): 1%. Though we’d love to endorse him, we can’t ignore the accusation that his name derives ultimately from the Latin ‘Benedictus’, reminding us too much of the Yiddish ‘bensching’. Sorry.

Amsalem, Haim (Am Shalem) – 3%. Though his name (אמסלם) is Arabic in form, his party’s name is a decent Hebraization (עם שלם). Will voters forgive? Probably not.

That’s all the candidates I can include without incurring the wrath of my editors. Hatzba’ah Tovah!

*Except the previous PM, Ehud Olmert, and as a result he was forced out of office.

Tuesday, 25 December 2012

Israeli Demographics: Jewish Minority?

If I had a penny for every time I saw someone warn that Israel is under threat from an Arab demographic time bomb, I would have collected a decent stack of pennies quite quickly, then stopped earning so much, as I stopped bothering to read the articles where you typically find this claim. The commonness of this claim, you might point out, could be be because it is true. Certainly, there are several studies which periodically predict a reduction of the Jewish majority in Israel. Looking at the current fertility rates - 2.98 children per woman for Jews, but 3.51 for Muslims (who make up the majority of Israeli Arabs) - they seem to support this claim. If you include the Palestinian Territories, with their fertility rate of 4.3, the nail seems to be in the coffin.

However, in 2006, an analyst and former Israeli ambassador, Yoram Ettinger, claimed a nail was missing, and the demographics were actually strongly in favour of the Jews. In conjunction with Bennet Zimmerman et al, he published a report asserting that the Palestinian Authority were claiming a population size 1 million larger than reality. He also noted that while the Jewish fertility rate was increasing, the Arab fertility rate was decreasing. Ettinger highlights alarmist reports from the 40s to the 90s which incorrectly predicted an Arab majority (either in pre-1967 Israel or in the entire area of former British Mandate Palestine) by such-and-such a time which has since passed. More recently, he goes so far as to predict an attainable "80% Jewish majority in the combined area of the West Bank and pre-1967 Israel" by 2035 (given increased Jewish immigration).

Ettinger's report flies in the face of conventional wisdom and has repeatedly been met with skepticism (in part because he is not a professional demographer - see bottom for the response of Sergio DellaPergola, commonly considered Israel's leading demographer). But is there any truth behind his claims? To examine the demographics, I will (wherever possible) use the most recent data from the Israeli Central Bureau of Statistics (CBS). 

Which demographic indicator is the most useful one to use - population growth rates, total fertility rates, birth rates (and death rates), life expectancy, and/or perhaps some other? Each is useful for different things. For long range population size, the most important indicator is fertility rates (see for example UN population estimates).

[NB: Life expectancy must also be taken into consideration, though in this case it is not necessary, since we are comparing two populations with similar life expectancy. There is a lag effect on the fertility rate: if one population is more youthful, as are the Arabs, it will take a while for a decreased fertility rate to have an effect on their relative population size, since the ageing population will have more deaths. Eventually the fertility rate will "win" as the average age stabilises. See here for the age distribution of Jews and here for the age distribution of Arabs.]

Crudely simplifying the contrasting claims, I would say that the main thrust of the demographic time bomb argument (henceforth, DTB) is that the Arab fertility rate is higher than the Jewish fertility rate and thus there will eventually be an Arab majority; while the main thrust of Ettinger's argument is that the gap between the fertility rates is narrowing and will eventually reverse, so there will eventually be a stronger Jewish majority. For now, we will leave migration out of things. The graph below shows the variation of each religion's fertility rate with time:


Source: CBS Statistical Abstract of Israel 2012
We will take the Arab fertility rate to be equivalent to the Muslim fertility rate, the green line, since Muslims make up the vast majority of Israel's Arab population. (In fact, it will be slightly lower than this, once merged with the Druze and Christian statistics.) The central statistical claim of the DTB argument is clearly factually accurate, and it is easy to see why the argument is so prevalent. For the entirety of Israel's history, the Muslim fertility rate (green) has been significantly higher than the Jewish fertility rate, and was as high as 9 children per mother in the 60s. It is also clear that Ettinger's central statistical claim is correct: the gap has significantly narrowed over the years. An interesting exercise is to compare the popularity of the DTB argument with the fertility rates. In the 90s, the argument gained widespread popularity with the ascendance of the concession-happy Left, and this coincides with the Muslim fertility rate remaining constant at roughly 4.7 for an extended period of time. Just before Ettinger wrote his report, in 2006, there was a sudden dip in this rate, and it has slowly but steadily decreased since then. Likewise, the Jewish fertility rate hit an all-time low in the mid-90s, and then gradually increased.

So far we have considered the demographics within sovereign Israel: that is, pre-1967 Israel, plus the Golan Heights and all of Jerusalem, adding the Israeli citizens who live in Judea and Samaria into the figures. In these areas, Jews constitute 75.4% of the population, and Arabs 20.5%. But the main political-demographic debate concerns whether Israel could extend sovereignty to the areas of British Mandate Palestine which it conquered in the Six Day War - Judea, Samaria and Gaza - and award citizenship to all its Arab residents, while still maintaining a long term Jewish majority. If we exclude Gaza (which is all but a separate country) from this, the proportion of Jews is... well, this still depends on whose figures you rely upon. Using the estimates in the CIA World Factbook, the Jewish population is 58% (Ettinger claims 66%). Crucially, however, the fertility rate in Judea and Samaria is 2.98 (according to the World Factbook) - exactly the same as Israel's Jewish population! Furthermore, this rate has steadily been falling

[NB: It is unclear whether the World Factbook figures for the population and fertility rate include the Jewish population in Judea and Samaria. In both cases I have assumed not. If they do, then the total Jewish population would be 61%, and the fertility rate for Arabs would be lower than 2.98 (since the Judea and Samaria Jewish fertility rate is above 5). Additionally, the World Factbook estimates the fertility rate within Israel as 2.67, which is lower than the CBS estimate. Since it is unlikely that the CBS are incorrect, it is possible that the World Factbook systematically underestimates fertility rates (or simply inaccurately estimates them). Other estimates for these figures are difficult to find, since most databases (UN, World Bank) pool the population statistics with those of Gaza.]

So who is right? The answer is not clear. The trouble with the DTB argument is that they are only considering the static fertility rates, and require that of the Arabs to remain high (or at least not change much) to be correct. This is problematic since the fertility gap has evidently been closing over the last few years. On the other hand, Ettinger's argument requires this gap to continue closing constantly, and ultimately reverse. There is no reason to assume the current trend will continue in the long term, particularly as there is no measured precedent of the Palestinian Arab fertility rate being lower than the Israeli Jewish fertility rate, though there are other Arab and Muslim countries which do have a lower fertility rate (Syria, Egypt and the Gulf States). He also excessively relies on fertility rates, which have a lagged effect, particularly due to the youthfulness of the Arab population.

Demographic projections are, of course, educated guesses. They are all based on models, and the line has to be drawn at some point regarding the complexity of the model. The DTB model considers the current population and its rate of change (sort of), while the Ettinger model also takes into account the rate of change of the rate of change (sort of). I have yet to see a model which adds in the 3rd order rate of change. There is no reason to assume that this (or the 4th, 5th etc. rate of change) will be zero, although assuming they have "reasonable" values, they take longer to have an effect. In any case, such complexity would be contrived. Before making an accurate long term model of the population, you'd have to make an accurate long term model of the entire universe, including the minds of men. 

So many other unpredictable factors affect Israel's population balance - principally, migration. In the past, the demographic balance has been saved by unexpectedly large immigration of Jews, such as those from the Former Soviet Union. Large scale Jewish immigration from France or the US would boost Ettinger's case, while greater emigration of Jewish Israelis, which usually occurs for economic reasons, would boost the DTB argument (every now and then, Haaretz print a victorious article declaring how many Jews want to leave Israel). The data themselves may even have some feedback influence on fertility rates - if Jews sense they are losing their majority, they may have more children (which may already be happening). Lastly, if you insist on including Gaza in all these calculations, with its third-world birth rate, Greater Israel seems a far more distant pipe dream.

If all these factors remain minimal, it boils down to how you extend the green and blue lines in the graph above, in your mind. If they cross, you are probably an evil right wing settler who destroys other people's olive trees before breakfast, and shifts your fenceposts 5 metres down the hill whenever nobody's looking. If they suddenly reverse you're most likely a starry-eyed liberal currently in India to find yourself; you have a strong sense of justice and aren't too concerned by factual reality. In my opinion, the truth is somewhere in between Ettinger and the DTB: Ettinger is excessively optimistic about the Jewish majority, which requires all of his claims to be true and remain true; but the DTB is excessively pessimistic given the trends. Ettinger's report does not allow Israel to annex Judea and Samaria without fear of demographic consequences. But it does demonstrate that there is more time to reach a calculated solution without giving in to demographic panic.

I don't expect people will take Ettinger's study into account much when writing their doom-and-gloom articles for a while, unless the fertility gap closes significantly more. Until then, I expect to continue collecting (hypothetical) pennies.

For further information, see a more recent article by Ettinger, interviews with Ettinger and Sergio DellaPergola where they dispute each others' claims, and a direct debate. Also see DellaPergola's very long and detailed report for the UN concerning the influences on Israel's fertility.

Friday, 26 October 2012

Likud-Yisrael Beiteinu Post-Merger Flash Poll

For anyone who hasn't seen, the two largest parties on the "right" (Likud and Yisrael Beiteinu) have just announced that they will run on the same electoral list. This helps vindicate my assertion a short while back that plenty could happen between then and the election, and plenty has. Anyway, the next item of news, apart from all the predictable shrill doomsday predictions of the party leaders on the "left", was that a quick-off-the-mark internet-based poll gave some preliminary results on how the electorate will vote in the new order, and it doesn't seem good for the merged list:

Panels Politics Poll, 25th October
 "Right"
"Centre-Left"
Haredi
Arab 
33 [42] Likud and Yisrael Beitenu
13 [07] Jewish Home and National Union
27 [08] Labor
18 [---] Yesh Atid
05 [03] Meretz
00 [28] Kadima
00 [05] Independence
09 [11] Shas
05 [05] UTJ
04 [04] Hadash
03 [04] Ra’am-Ta’al
03 [03] Balad
Total:        46
50
14
10
Cheers, Knesset Jeremy.

Instead of (what was presumably) the intended aim of increasing their number of seats, the joint list has actually gone down by 9 seats. According to this poll, the biggest winners are the Religious Zionist parties (Jewish Home and National Union), Labor, and Yesh Atid. The biggest losers are the Likud-YB merger, Kadima and Independence (though the last two have been polling very low anyway).

However, all is not necessarily what it seems. It is my belief that this poll is flawed in several ways. That is not to say that its conclusions are wrong, they may well be right; but there is not yet enough clean data to strongly support its conclusions.

Firstly, looking at the sample size of the poll, 305 people were asked. That's good for such an off-the-cuff poll, but the normal sample size in these polls is about 500 (which leads to a 4-5 seat error). The percentage error is given by the rough formula: %error = 100 / sample size (for those of you who care even more, that's the error at the 95% confidence level). Thus, in this poll the error is more like 7 seats, which is fairly big. As an example, the 33 seats the poll awards to the merged party could really be anything between 26 and 40, and there's a small chance (1 in 20) that it's actually outside that range.

So far we've only considered random errors. These assume that the poll was well-meaning, but just by bad luck happened to pick more Labor supporters than Likud fans. However, more damningly, there may also be systematic errors - the kind you would get by conducting your poll in Umm Al-Fahm or Mea Shearim, likely to give responses unrepresentative of the country. Here I will be more speculative, and you may disagree with my assertions, and you may also think of other systematic errors I haven't noticed.

The facts are as follows: the poll was conducted on the internet, after 8pm, immediately after the announcement. Alarm bells should be ringing. If any of you have visited the internet, you'll know it's a favourite haunt of some points of view. The point is, I suspect that "left wing" people spend more of their time on the internet, and so were oversampled in this poll. Justification: in Israel it's not just socioeconomic issues that dominate, and so while a large base of support for "left wing" parties like Labor lies in affluent Tel Aviv and the Dan bloc, a large base of support for the Likud is in the development towns, and they may spend less time on the internet (though internet usage in Israel is high). Please note, I'm not saying this is true: I haven't based it on any evidence; I'm saying it's a possibility, which may call into question the validity of the poll.

Additionally, since the poll was conducted so soon after the event, it's possible people got their first source of information from the poll's question itself, which begs the question of the wording of the question (if you'll pardon the pun). Lastly, and perhaps most importantly, people's opinions are not formed by their own independent thinking alone. The day after the event, people will chat about it at work, and commentators will write about it in newspapers, and people's opinions will be formed (in part) by cognitive dissonance or persuasion by these views.

Please let me know if you think of any more issues, or why the above are not problematic! In conclusion, the poll is not (necessarily) representative of the election outcome, and possibly not representative of immediate reactions to the announcement either. Future standard polls will indicate the real effect of this.

Friday, 19 October 2012

Election Polls #1

The best place to get all the polls from (and I hesitated to reveal my excellent source) is "Knesset Jeremy" (KJ), an absolute don who publishes more or less every electoral poll taken in Israel, though sometimes late. Huge respect to the man. At the moment polls are being pumped out faster than the water in the Kinneret, so there's lots to do. I present here some of the most recent polls, in a slightly different format from KJ. I have put together parties which I think would naturally form coalitions together. The Haredi parties (particularly Shas, though perhaps no more) may be more inclined to join the "right" in government. 

I have not included scenario polls here, but only polls based on the current set of parties standing for elections. If you do look at the scenarios, note that none of the potential changes seems able to change significantly the number of seats the "right" and "left" each get. Since these polls were conducted, Aryeh Deri has joined Shas, which is expected to increase their number of seats at the expense of the Likud. Sometimes the polls will group together the two Religious Zionist parties (who have announced that they intend to run on one list), and the Arab parties (who are not). 

Dates indicate when KJ uploaded the poll. The numbers in square brackets indicate the number of seats the party currently has in Knesset, with the other number (obviously) giving the projected number of seats. 


Globes Smiths Poll, 19th October
 "Right"
"Centre-Left"
Haredi
Arab 
27 [27] Likud
15 [15] Yisrael Beitenu
05 [03] Jewish Home
04 [04] National Union
18 [08] Labor
14 [---] Yesh Atid
04 [03] Meretz
03 [28] Kadima
02 [05] Independence
11 [11] Shas
06 [05] UTJ
04 [04] Hadash
04 [04] Ra’am-Ta’al
03 [03] Balad
Total:        51
41
17
11


Haaretz Dialog Poll, 18th October
 "Right"
"Centre-Left"
Haredi
Arab 
29 [27] Likud
13 [15] Yisrael Beitenu
06 [07] National Union and Jewish Home
20 [08] Labor
11 [--] Yesh Atid
09 [28] Kadima
04 [03] Meretz
00 [05] Independence
11 [11] Shas
06 [05] UTJ
04 [04] Hadash
04 [04] Ra’am-Ta’al
03 [03] Balad
Total:        48
44
17
11


Smith Poll, 18th October
 "Right"
"Centre-Left"
Haredi
Arab 
26 [27] Likud
13 [15] Yisrael Beitenu
10 [07] Jewish Home and National Union

22 [08] Labor
12 [--] Yesh Atid
06 [28] Kadima
03 [03] Meretz
00 [05] Independence
11 [11] Shas
06 [05] UTJ
11 [11] Hadash, Ra’am-Ta’al, Balad
Total:        49
43
17
11


Walla-Teleseker Poll, 18th October
 "Right"
"Centre-Left"
Haredi
Arab 
30 [27] Likud
12 [15] Yisrael Beitenu
05 [03] Jewish Home
02 [04] National Union
19 [08] Labor
14 [--] Yesh Atid
06 [28] Kadima
05 [03] Meretz
01 [05] Independence
09 [11] Shas
06 [05] UTJ
04 [04] Ra’am-Ta’al
04 [04] Hadash
03 [03] Balad
Total:        49
45
15
11


Thursday, 18 October 2012

Shas

I said in my last post that anything can happen, and there have been a number of developments since I wrote that. I want to focus on the changes in one particular party: Shas, which stands for "Shomrei Sefarad", the "guardians" of "Sefarad" (here, the Sephardi world; and the guardians being the guardians of Torah, according to Wikipedia). This is a Sephardic Haredi party which was founded in 1984 with the purpose of bringing greater equality to Jews from Arab lands, who had faced unofficial discrimination in the first few decades of the State and faced difficulties reaching the higher echelons of society. In the previous decade, the Likud party had been the main hope for these Jews.

Though the party's official ideology is Haredi and guided by the Shas "mentor" Rav Ovadia Yosef, large numbers of secular Sephardi Jews vote for them, possibly out of a historical connection to this party which fought for their rights, or possibly because (as some say) they believe deep down (in spite of their secular lifestyles) that the party's ideology is true, and express this through their votes. This explains their large number of seats (11 of 120 in the current Knesset), comprising almost 10% of the Knesset, whereas Sephardi Haredim are nowhere near 10% of the country's population (thought to be 20% of Israel's Haredim which in turn are thought to be just over 10% of the country; though many demographers think Haredim are under-counted). In 2010, Shas joined the World Zionist Organisation and now describes itself as a Zionist party, in contrast to the Ashkenazi Haredi party (United Torah Judaism) who do not.

The current Shas party is considered "right wing", which really refers to their nationalistic and hawkish character. The leader, Eli Yishai, is thought to drive this trend, and the party is currently in the ("right wing") governing coalition. However, during the 90s, the party aligned more with the left, and was led by Aryeh Deri, who was convicted for corruption and served time in jail. He has now finished his refractory period after being released from prison, and is allowed to enter politics again. He was promised by Rav Ovadia Yosef that he would return as party leader; a compromise deal has just been struck with Eli Yishai (current leader) whose details are not entirely clear. What we know is that Yishai will get the first place in the list and the first ministerial slot offered to the party if they join a governing coalition, and Deri will get the second place, and will lead the election campaign; neither will be the party leader. The logic behind this is that Deri is considered the more charismatic leader and so will attract more votes.

The reason I want to mention all of this is that Shas is projected to attain 10-15 seats by the polls, and with that many seats, they could align with either the "right" or the "left" and give them a majority. Depending on who has more power between Yishai and Deri, this may decide the outcome of the election. (Judging by the polls, none of the scenarios yet conceived, such as new parties being created, have actually been able to win a significant number of voters from the "right" to the "left". An existing party changing its alignment seems, currently, the only way the "left" could form a coalition.)

My speculation is as follows: If Shas join a coalition, they will surely demand a ministerial position, which automatically (according to the deal) goes to Yishai. I would be surprised if the "left" would be willing to allow Yishai (with his "right wing" views) to hold any serious positions. Additionally, one of the main concerns of Shas at present is the impending conscription of the Haredi sector to the army, and drafting a law to allow some to continue learning Torah full-time. This, I believe, is the clincher - Shas are highly unlikely to think that they will get a better deal on this from the "left" than the "right", particularly given many of the statements by "centre-left" leaders on drafting everyone into the army. That said, Shas managed to be in a coalition with Yisrael Beiteinu, who also insist on equal conscription; though media reports suggested that the coalition was straining towards the end partly over this issue. I find it even less likely that (if they join a "right wing" coalition) they would oppose key "right wing" issues such as settlement expansion, and risk losing the support of the Religious Zionist parties on issues of religion and state.

Another Shas-related development is that MK Rabbi Haim Amsalem, who broke off with Shas a couple of years ago to form his own party, "Am Shalem" (whole nation), which peddles a more Modern-Orthodox stance on several issues, has finally announced that the party will be running for election on its own list. Polls are inconclusive about whether his party will cross the election threshold.