Graphs & Data

v  Thirty-eight percent of the number of housing loans issued by the banks between 2010 and 2013, and about half their volume were issued to the three highest income deciles. Fifty percent of the number and 43 percent of the volume of housing loans were issued to the middle deciles, and 13 percent of the number and 7.5 percent of the volume were issued to the three lowest deciles.
v  The number of monthly salaries required to finance the purchase of a home through a mortgage to borrowers in the lowest quintile is 130—twice as high as for borrowers from the highest quintile.
v  The average payment-to-income ratio (PTI) of housing loans to borrowers from the lowest quintile is 40 percent, compared to an average rate of 30 percent in the highest quintile.
v  The loan-to-value ratio (LTV) is distributed similarly across all income deciles, and declined from an average of 54 percent in 2010 to an average of 52 percent in 2013.
 
This box presents the main characteristics of housing loans issued by the seven largest banks between January 2010 and December 2013, constituting 58 percent of outstanding housing credit at the end of 2014 (about 309 thousand loans totaling about NIS 153 billion).  Data on these loans serve the Banking Supervision Department in its analysis of housing credit as part of the macroeconomic stress test, based on a bottom-up scenario, that it conducts on the banking system.
 
The main data obtained from a breakdown by income levels
The individual data on borrowers provide information on household income[1], and are classified into (net) income deciles in accordance with the income survey on households with a salaried head of household that was conducted by the Central Bureau of Statistics in 2011.[2]
An examination of how the number (Figure 1) and volume (Figure 2) of loans is distributed among the income deciles shows that the three highest deciles received about 38 percent of the loans during the reviewed period, and the volume of these loans accounted for about half of housing credit.  The middle (fourth through seventh) deciles received 50 percent of the loans, accounting for 43 percent of housing credit, and the three lowest deciles received 13 percent of the loans, accounting for 7.5 percent of the volume of housing credit.  We can also see that this distribution was brought into sharper relief over the period between 2010 and 2013.  More loans were issued to the higher deciles in 2013, both in terms of the number of loans and in terms of their volume, and the lowest deciles received a very small share of the total number of housing loans and of the volume.
In 2013, 11 percent of the number of loans issued by the banks helped borrowers purchase a home for investment.  Most homes for investment that were financed in that year through mortgages were purchased by buyers in the three highest deciles (close to 8 percent of all loans—Figure 3).  Nineteen percent of the number of loans issued to the three highest deciles in 2013 were for the purchase of homes for investment, while in the three lowest deciles, 4 percent of the number of loans were for the purchase of a home for investment.
10-06-2015-E1.png

 10-06-2015-E2.png
10-06-2015-E3.png
 
The average number of monthly salaries necessary for a household from the lowest quintile[3] to purchase the home they purchased with a mortgage—about 130—is twice as high as the number of monthly salaries necessary for a household from the highest quintile—about 65 (see Figure 4).  Figure 4 also shows that during the reviewed period, there was an increase in the number of monthly salaries necessary to purchase a home, particularly among the lowest deciles.  The data show that the debt burden increases with a decline in income (Figure 5).  In the highest quintile the PTI ratio was an average of about 30 percent during the period, while in the lowest quintile, the rate was close to 40 percent.  In terms of the risk level of the loans (Figure 6), the percentage of high risk borrowers declines with an increase in income decile, and the percentage of low-risk borrowers increases.
The lowest deciles are therefore characterized by a high payment to income ratio, and the number of monthly salaries necessary to buy a home is much higher than the number of monthly salaries necessary for the highest deciles.  The burden that they take upon themselves is therefore higher than the burden taken on by the highest deciles, and the same is true for their risk level.
10-06-2015-E4.png

10-06-2015-E5.png
10-06-2015-E6.png


An examination of how the prices of homes financed by a mortgage are distributed by income deciles (Figure 7) shows that there is a positive correlation between home price and the borrower’s income level, and that the variance increases with the income level.  A similar picture is obtained when examining how the loan level is distributed within each decile (Figure 8).  Since the LTV ratio of the loans is equal to the ratio between the approved framework of the loan and the value of the asset, it is distributed equally in all income deciles (although there is larger variance in the lowest decile; see Figure 9).[4]  Furthermore, the average LTV ratio in the entire sample declined over the period from 54 percent to 52 percent.
 
10-06-2015-E7.png

 10-06-2015-E8.png
10-06-2015-E9.png


The main data obtained from a breakdown by geographic regions
 
The individual data on borrowers provides information on the location of the asset.[5]  We classified them by region according to the Central Bureau of Statistics definition of regions.[6]
An analysis by region shows that the highest home prices were registered in Tel Aviv, and in the surveyed period, they averaged NIS 2.23 million per home (Figure 10).  Relatively high prices were also estimated in the Center, Sharon, Gush Dan and Jerusalem regions.  The payment to income ratio is distributed similarly by region (Figure 11).  The highest figure was recorded in the Tel Aviv region (36 percent), followed by the Center, Sharon, Gush Dan and Jerusalem regions, which ranged from 31 percent to 33 percent).  The highest LTV ratio was recorded in the South, Krayot, North and Haifa regions, with the average ranging from 58 percent to 59 percent (Figure 12).  In the rest of the regions, the average LTV ratio was slightly lower, ranging from 53 percent and 54 percent.
10-06-2015-E10.png
10-06-2015-E11.png


10-06-2015-E12.png


[1] Total income for the purpose of calculating the mortgage repayment ratio—includes net monthly income plus other income/expenditure for the purpose of calculating income.
[2] The Central Bureau of Statistics uses net monthly household income.  The deciles according to income ranges are: the lowest decile—up to NIS 5,174; the second decile—between NIS 5,174 and NIS 6,811; … and the upper decile above NIS 25,408.
[3] The lowest quintile includes the two lowest deciles.  The second quintile includes the third and fourth deciles, and so forth.
[4] The LTV ratio provides an indication of the level of leverage of those taking out mortgages.
[5] As opposed to the location in which the mortgage was taken out.
[6] In order to examine the housing field, the Central Bureau of Statistics defines nine regions: South, North, Haifa, Krayot, Jerusalem, Tel Aviv, Gush Dan, Sharon and Center.  In order to examine the fields of income, unemployment and so forth, the Central Bureau of Statistics uses a slightly different definition of regions.