SANTA MONICA RENT CONTROL BOARD www.smgov.net/rentcontrol The Impact of Market-Rate Vacancy Increases Twelfth Year Report 1999-2010 Stephen Lewis, Public Information Manager 03/10/2011 A summary, with analysis, of Costa-Hawkins vacancy increases in the twelve years from January 1, 1999 to December 31, 2010.
Introduction In the late 1970s, demand for Santa Monica apartments outpaced supply (due in part to demolition of apartment buildings), leading to a sharp increase in rents and threatening broad economic dislocation. In response, the city s voters amended the city charter to create the Rent Control Law. The law worked as the voters intended until 1995, when the Legislature in Sacramento enacted the Costa-Hawkins Rental Housing Act, limiting cities authority to respond to local housing conditions by means of permanent rent control. Under this state law, a unit s rent is decontrolled at the end of a tenancy. The owner can set a new rent for the next tenancy which is then controlled at that level, leading some to characterize Costa- Hawkins as a system of vacancy decontrol-recontrol. This report surveys Costa Hawkins s effect over the 12 years since it went fully into effect in 1999. The facts in this report speak for themselves, but a few are especially noteworthy: While the cumulative rate of inflation in southern California since 1999 has been 39%, median rents for decontrolledrecontrolled apartments have gone up by nearly 150% over the same period (see p. 6). Although market rents have fallen slightly over the past two years (p. 4), a family of four still needs an annual income of $84,463 exceeding the area median income of $63,000 by more than $20,000 in order to pay the median rent on a one-bedroom apartment without undue burden (pp. 12-14). Since the decontrol-recontrol system began, 59% of controlled units have received vacancy increases (p. 3). What these facts mean for our community in the long term is not yet clear, but some things can be hypothesized. It seems likely, for example, that many young people earning entry-level pay will be unable to afford to live in Santa Monica, and those living on fixed incomes principally seniors and the disabled will likely be unable to continue to live here if they lose their long-term rentcontrolled housing. If true, these things cannot help but affect our community s culture and diversity, both economically and more generally. It is our hope that the release of the 2010 Census data will allow us to analyze these issues in next year s report. Tracy Condon, Administrator March 2011 Page 1
Costa-Hawkins: a Two-Tiered System There are two broad types of rent control. The first type, strict control, establishes a base rent and then controls the rent at that level, usually adjusted annually for inflation, more or less in perpetuity. The rent will remain at the same inflation-adjusted level no matter how many tenants come and go, because the rent level runs with the unit, not with the tenancy. The second type, called vacancy decontrol (or decontrolrecontrol) establishes a base rent at the beginning of each new tenancy, and controls the rent at that level again, adjusted annually for inflation for the duration of that tenancy only. The control ends when the tenancy ends, and the owner can charge the next tenant whatever rent he or she can negotiate in the open market. The rent is then controlled at that level, plus inflationary adjustments, until that tenancy ends and so on. When the people of Santa Monica enacted rent control by referendum in 1978, they opted for a strict-control system. Similar strict-control laws were enacted at about the same time in Berkeley, West Hollywood, and San Francisco, and later by East Palo Alto. Other jurisdictions, notably Los Angeles, enacted vacancy decontrol-recontrol rent-stabilization laws. With the passage of the Costa-Hawkins Rental Housing Act in 1995, strict rent control was replaced by vacancy decontrolrecontrol in every rent-control jurisdiction statewide. The replacement was phased in, however, rather than being imposed immediately. Those tenants whose tenancies began before 1999 would continue to be subject to strict rent control; so even tenants who moved in after Costa-Hawkins s passage, but before 1999, would have their rent controlled with reference to the base rent established when the local rent-control law was established, and their rent would remain controlled at that level (with annual adjustments based on increases in owner s operating costs) for the duration of their tenancy. But the rent for any tenancy beginning after January 1, 1999 would be set at whatever rate the owner negotiated with the new tenant, and would be controlled based on that initial market-rate level for the tenancy s duration. The result 12 years later is that there is a two-tiered rent-control system. There is a still-significant but shrinking minority of tenants whose rents are based on what was charged in 1978, and an expanding majority whose rents are based on market conditions Page 2
existing when their tenancy began. Thus, there is a growing distinction between rents for long-term controlled units and market-rate units, even though both are subject to the same rentcontrol law. Unit size Long-Term Controlled Decontrolled- Recontrolled No Registered MAR % of total units Median MAR number of units % of total units this size that received no vacancy increase Median MAR number of units % of total units this size that received vacancy increases % of all vacancy increases number of units % of total units with no registered MAR Total number of units this size 0 bedrooms 11% $600 854 27.5% $1,140 2,070 66.6% 12% 182 5.9% 3,106 1 bedroom 47% $755 4,775 36.4 % $1,525 8,328 63.4 % 50% 33 <1% 13,136 2 bedroom 34.5% $946 4,199 43.4 % $2,006 5,372 55.5% 32% 100 1.1% 9,671 3+ bedroom 7.5% $1,186 1,179 55.9% $2,675 862 40.8% 5% TOTALS 11,007 16,632 70 3.3% 2,111 (11,007 + 16,632 = 27,639 actively controlled units) 385 100% 39.3% received no vacancy increase 59.3% received vacancy increases 1.4% with no registered MAR 28,024 Page 3
As the table on the previous page shows, there are 28,024 units currently subject to the Santa Monica Rent Control Law. 1 As of December 31, 2010, 39% of those units were occupied by longterm tenants and 59% were occupied by market-rate tenants. 2 The median rents for decontrolled-recontrolled units (shown in the middle column) are significantly higher than those for long-term controlled units (shown in the second column). The Economy s Effect on Market Rents While market rents are higher than long-term controlled rents (roughly double, on average), much of the difference results from sharp increases in rents in the decade between January 1999 and December 2008. Median rents for newly-rented units actually declined over the past two years, falling to levels not seen since 2005-2006. The graph below (with a related table at the top of the next page) illustrates the rise and fall at a glance. Median Vacancy-Increase MARs from 1999 2010 Vertical axis = $US. Horizontal axis = year 1 Excluded from this total are 8,557 units that have either been removed from rent control or are currently subject to various use exemtions (1,466 are owner occupied; 1,953 have been Ellised; 1,639 have removal permits; 756 are nonrentals or commercial; and 2,743 are subject to various other use exemptions. 2 385 units have no MAR for varying reasons, including that: they have been owner occupied since 4/10/79; they are not rented; they are unregistered. Page 4
Median vacancy-increase rent by year (in $US), broken down by bedroom size 99 00 01 02 03 04 05 06 07 08 09 10 0 BR 800 850 895 925 961 985 1,050 1,164 1,250 1,295 1,223 1,195 1 BR 1,000 1,175 1,225 1,239 1,250 1,300 1,358 1,495 1,595 1,645 1,525 1,495 2 BR 1,400 1,600 1,695 1,620 1,679 1,775 1,850 1,995 2,150 2,200 2,075 2,000 3 BR 1,800 2,075 2,100 2,250 2,300 2,400 2,600 2,940 2,979 2,987 2,775 2,691 This decline, with rents falling to a level seen in the middle of the last decade, reflects the general economic downturn, with rents reaching their peak in 2007-2008 and declining steadily thereafter. 3 Viewed in isolation, the decline in market-level rents may seem dramatic. But focusing on the decline alone paints a somewhat misleading picture; it didn t occur in a historical vacuum, after all, but in a broad economic context. And within that context the change in market level rents over the course of more than a decade the decline s significance pales considerably. Again, median rents have approximately doubled from before Costa- Hawkins s full implementation in 1999 through the end of last year, even with the recent decline. The significance of the 2009 2010 decline in rents pales even further when measured against the economy as a whole. From December 1998 to December 2010, southern California s cumulative inflation was 39%. 4 The average inflation in market-rate rents over the same period was nearly 150%, exceeding overall inflation by just under 110%. The Two-Tiered System in Context Many of our sister rent-control jurisdictions allow annual general adjustments (the amount by which rents are adjusted for inflation) that are a percentage of the local general inflation rate. The City of Berkeley, for example, currently allows general adjustments equal to 65% of the rate of increase in the Consumer Price Index (CPI) 3 The recession officially began in December 2007. Casey Mulligan, What Caused the Recession of 2008? Hints from Labor Productivity, National Bureau of Economic Research (NBER Working Paper No. 14729, Issued in February 2009) 4 Source: United Stated Bureau of Labor Statistics (CPI-U [all items, all urban consumers] for Los-Angeles, Riverside, Orange County) Page 5
over the preceding year. The City of West Hollywood allows general adjustments equal to 75% of the increase in the CPI. Since Costa-Hawkins s full implementation in 1999, the CPI for the Southern California region that includes Santa Monica increased by 39%. Applying Berkeley s current 65% formula would have yielded a cumulative increase in a continuously-controlled rent over the same period of 25.35%. The West Hollywood formula would yield an increase over that period of 29.25%. Santa Monica s cumulative general adjustment for long-term controlled units over that period has been a relatively robust 34%, about 80% of the general inflation rate. Because of Costa-Hawkins, it is only units with uninterrupted tenancies from before 1999 until now that would be subject to those limitations. The rent for any new tenancy may be set at whatever level the market will bear, allowing for deviations from what would otherwise have been a uniform 34% cumulative increase. The inflation in median rents for 0-bedroom units subject to Costa-Hawkins vacancy increases has been 115% since the end of 1998. Median rents for larger units have inflated by 167%. Unit size by number of bedrooms (with Median MARs on 12/31/1998, before vacancy decontrol) Average % Increase in median MARs from 12/1998 to 12/2010 if there had been no vacancy decontrolrecontrol What median MARS would be today without vacancy decontrolrecontrol Actual Median MARs of vacancy decontrolledrecontrolled units % increase in actual median decontrolled recontrolled MARs from 12/1998 to 12/2010 Area inflation rate (Change in CPI-all items-for Los Angeles, Riverside, Orange County) December 1998 December 2010 Percent by which vacancy decontrol recontrolled rents exceed the rate of inflation 0 ($531) 34% $711 $1,140 115% +76% 1 ($611) 34% $815 $1,525 150% +111% 2 ($778) 34% $1,044 $2,006 158% +119% 39% 3 ($1,002) 34% $1,316 $2,675 167% +128% Page 6
An Increasing Disparity? It is inevitable that there will come a time when all controlled units will have been rented at market level at least once and there will be no long-term tenancies left. On the way to that happening, it seems likely that the disparity between long-term and marketrate tenancies will only increase, the recent decline in market rents notwithstanding. This seems likely for two reasons. First, the fact that a unit experiences at least one vacancy increase makes it likely that it will receive another such increase in a relatively short period of time. This is intuitively reasonable: tenants in long-term controlled units, whose rents are well below market level, have a powerful economic incentive to remain in place; those whose rent is similar to (and in some cases more than) the amount being charged for a comparable unit in the same neighborhood do not. It is also borne out by the evidence. Of the 16,632 units that have experienced a Costa-Hawkins vacancy increase, 32.7% have experienced only one. The remaining 67.3% have experienced two or more (25.4% have been rerented 4 or more times since 1998; 17.6% have been rerented 3 times; and 24.3% have been rerented twice). Second, a decline in the number of long-term controlled rents is inevitable no matter how affordable they are. Tenancies end for reasons having nothing to do with cost: people move to new areas for work or other opportunities, or to be closer to family; they die or become too ill to remain in their homes; and single people fall in love and move in together, leading one to give up his or her previous residence. As this process continues, long-term controlled tenancies will become an ever smaller share of the overall controlled housing stock. Those affordable units are likely to be held on to more tightly as their more-affordable rent becomes an increasingly better deal. And as the table at the top of the next page shows, that is exactly what has happened. Page 7
To allow closer analysis, the following table shows the degree of turnover in each city area (by number and percentage) broken down by bedroom size, along with median rents for long-termcontrolled and decontrolled-recontrolled units of each size in each area. In the early 1990s, the Board divided the city into 7 areas corresponding to neighborhoods and census tracts. Each area was designated by a letter, A through G. F G Number of units in category Percentage of units in category E C D A B Unit size AREA A 0 bedrooms $560 129 28% 1 bedroom $784 966 36% 2 bedrooms $982 615 44% 3+ bedrooms $1,157 153 65% AREA B AREA C AREA D No Vacancy Increase 0 bedrooms $563 117 30% 1 bedroom $702 640 39% 2 bedrooms $844 490 45% 3+ bedrooms $1,075 126 52% 0 bedrooms $658 93 22% 1 bedroom $765 213 50% 2 bedrooms $2,341 154 57% 3+ bedrooms -- -- --% 0 bedrooms $465 64 34% 1 bedroom $665 512 40% 2 bedrooms $732 606 51% 3+ bedrooms $894 157 62% Vacancy increase No registered MAR Total units this size $1,150 304 67% 22 5% 455 $1,650 1,746 64% 6 <1% 2,718 $2,233 774 55% 11 <1% 1,400 $2,417 80 34% 4 >1% 237 $984 254 64% 24 6% 395 $1,349 998 61% 4 <1% 1,642 $1,842 600 55% 9 <1% 1,099 $2,216 105 43% 12 5% 243 $1,515 332 77% 7 >1% 432 $2,162 212 50% 1 <1% 426 $8,094 90 33% 27 10% 271 $12,119 30 97% 1 3% 31 $1,000 77 41% 45 24% 186 $1,298 754 60% 1 <1% 1,267 $1,636 554 47% 19 2% 1,179 $2,279 84 33% 12 5% 253 Page 8
Area Difference in median MARs (averaged for all bedroom sizes) between units with and without vacancy increase A + 114% B + 100% C +535 % D +125% E +94% F +110% G +112% AREA E 0 bedrooms $602 128 20% 1 bedroom $713 741 34% 2 bedrooms $915 780 42% 3+ bedrooms $1,203 286 58% AREA F 0 bedrooms $676 191 34% 1 bedroom $852 782 36% 2 bedrooms $1,128 694 45% 3+ bedrooms $1,262 116 56% AREA G 0 bedrooms $600 132 30% 1 bedroom $755 921 34% 2 bedrooms $1,029 860 37% 3+ bedrooms $1,322 341 53% $1,071 472 72% 35 5% 635 $1,377 1,430 66% 7 <1% 2,178 $1,840 1,083 58% 14 <1% 1,877 $2,362 205 41% 6 >1% 497 $1,218 352 63% 13 >2% 556 $1,729 1,410 64% 4 <1% 2,196 $2,281 831 54% 8 <1% 1,533 $3,097 82 40% 8 4% 206 $1,199 279 62% 36 8% 447 $1,595 1,778 66% 10 <1% 2,709 $2,150 1,440 62% 12 1% 2,312 $2,913 276 43% 27 4% 644 A Closer Look Facts without context are meaningless at best. Worse, they are often misleading. Picture a close-up photograph of a man running. Is he catching a bus? Late for work? Just getting some exercise? But imagine expanding the frame and seeing that he is running to save a small child from the path of a moving car. There still would be things that we would not know (the man s relationship to the child, and how fast the car is traveling and so how great the danger), but with the context, we d know more than we knew before. Pull the frame out some more and see that the street is a film studio lot, and a cameraperson is filming the scene presided over by a director. Now we know ever more. So far, this report has examined Costa-Hawkins s effect since its full implementation in 1999 in order to provide context. That the Page 9
median rent for a one-bedroom apartment rented last year was $1,495 means very little by itself. The fact becomes more meaningful, though, when we know that the median rent for onebedroom apartments rented at any time over the last 12 years is $1,525. The new fact tells us that there was a decline in market rents at some point over that 12-year period. A look at the table at the top of page 5, which details rents year by year since 1999 tells us that the decline took place over the last two years. Together, these facts tell us that there has been a recent decline in market rents. Knowing that, even with the decline, median rents still are inflated well over 100% from their 1998 levels while inflation generally has been only 39% puts the decline into perspective. That perspective provides a context for the following detailed look at Costa-Hawkins s effect over the last one to three years. The below table shows the number of units that were given vacancy increases in 2010 in each area, broken down by bedroom size. The bar graph at the top of page 11 shows the median rent in each area, again broken down by bedroom size. Area C is excluded from both the graph and the table only to be consistent with past years reports, in which the area was excluded due to an unusual number of vacancies occurring in a small number of very large buildings, resulting in misleading statistics for that area. Bedrooms Area A Area B Area D Area E Area F Area G Totals by BR Size 0 61 42 9 63 76 52 303 1 308 133 184 232 262 279 1,398 2 120 89 77 162 139 276 863 3 9 18 13 28 6 44 118 Area totals 498 282 283 485 483 651 2,682 Page 10
Median MARs of Units with Vacancy Increases in 2010, by City Area A very small number of units were vacated and rerented more than once over the last year. In those cases, only the last Costa- Hawkins vacancy increase was used in the calculating the median MARs included on these graphs. The following table and the graph on the next page provide the same type of information as the preceding ones (also with Area C excluded for consistency), but for the past three years, rather than for last year alone. Bedrooms Area A Area B Area D Area E Area F Area G Totals by BR Size 0 136 124 31 201 201 125 818 1 918 425 410 646 704 806 3909 2 352 258 236 474 424 658 2402 3 28 40 27 86 34 136 351 Area Totals 1434 847 704 1407 1363 1725 7480 Page 11
Costa-Hawkins s Effect on Affordability That decontrolled-recontrolled rents are on average more than double those of long-term controlled rents is interesting in its own right because it tells us Costa-Hawkins s effect on rents. But what it does not tell us is how the law has affected the community. A roughly 150% increase in decontrolled-recontrolled rents may be a boon for owners and irritant for renters, but if it is nothing more than that, it is a matter of little public concern. The evidence shows that it is a great deal more. Median rents have not simply risen; they have risen out of the reach of many renters. Take the case of a single person earning minimum wage, which in California is $8.00 per hour. Annual gross wages based on a 40- hour work week and 52-week year at this rate total $16,640. Page 12
Under HUD regulations, income guidelines are adjusted for household size so that larger households have higher income limits than smaller ones, with 4- person households being used as a baseline. See Appendix, a memorandum from the California Department of Housing and Community Development. The median rent for a 0-bedroom long-term controlled apartment in Santa Monica is $711 per month. Under housing affordability standards, under which only 1/3 of household income (adjusted by a household adjustment factor see sidebar on page 13) should go toward housing expenses, a person would need to earn $40,629 a year in order to afford that rent without being rent burdened. To earn that much, the minimum-wage earner would need to increase his or her hours significantly, from 40 hours a week to 98. This would require him or her to work 14 hours a day, seven days a week. Thus, even a long-term controlled 0-bedroom apartment in Santa Monica is not affordable to a person earning minimum wage and working 40 hours a week. Such an apartment is, nonetheless, attainable if instead of using roughly a third of his or her income for rent, the worker dedicates 51% to it. With an annual rent of $8,532, the minimum-wage earner working full time at 40 hours a week will be left with $8,108 per year ($675 per month, or just under $170 per week) for the remaining necessities of life. It would be a difficult existence, and would be substantially complicated by things like illness or reduced hours that would lower the worker s income, but it is an existence that is at least possible as a practical matter. But this is not true with respect to a decontrolled-recontrolled 0- bedroom apartment, for which the median rent is $1,140 per month. For that rent to be affordable, the minimum-wage earner would need to work 8,143 hours per year, or a little more than 22 hours per day, 365 days per year. This would leave less than two hours each day to commute to and from work and sleep, and is obviously outside the bounds of human possibility. Unlike the longterm controlled rent, however, a 0-bedroom unit that has received at least one vacancy increase is not merely unaffordable as Affordability of units with no vacancy increase Unit Size Median MAR without Vacancy Increase household adjustment factor 30% 12 Annual income needed to afford Median MAR Area median income by household size Difference 0 $711/month.7.3 12 $40,629 $44,100 $3,471 1 $815/month.8.3 12 $40,750 $47,250* $6,500 2 $1,044/month.95.3 12 $43,958 $50,400 $6,442 3+ $1,316/month 1.085.3 12 $48,516 $63,000 $14,844 *Based on average of area median incomes of 1 and 2 person households Based on average of area median incomes of 3 and 4 person households Page 13
Affordability of units with vacancy increase Unit Size Median MAR with Vacancy Increase defined by HUD; it is unattainable. At an annual rent of $13,680, the median MAR for a 0-bedroom apartment would use a little more than 82% of the worker s pay, leaving $2,960 per year for the remaining necessities of life. Thus, the worker would have about $8 per day for such things as gas or bus fare, food, utility payments, health insurance, and clothing. Any unplanned-for event such as a needed car repair, illness, or reduction in hours would reduce even further the $8 per day on which the worker would have to live. household adjustment factor 30% 12 Annual income needed to afford Median MAR Area median income by household size Difference 0 $1,140/month.7.3 12 $65,143 $44,100 $21,043 1 $1,525/month.8.3 12 $76,250 $47,250* $29,000 2 $2,006/month.95.3 12 $84,463 $50,400 $34,063 3+ $2,643/month 1.085.3 12 $97,438 $63,000 $34,438 *Based on average of area median incomes of 1 and 2 person households Based on average of area median incomes of 3 and 3 person households Even those who can find a way to pay the higher rents are often perhaps usually significantly rent burdened, or must share housing with other wage earners. As the table on the preceding page shows, it is not only minimum-wage earners who are effectively priced out of the Santa Monica rental housing market for decontrolled-recontrolled units. At every household level and corresponding unit size, the median MAR for a Santa Monica apartment with a vacancy increase is tens of thousands of dollars more than what is affordable under HUD guidelines. This is not true for long-term controlled units. As the table on page 12 shows, those units remain near, but within, the affordability level for all household sizes. The table at the top of page 14 compares the number (and percentage) of units affordable to tenants whose incomes were very low, low, moderate, and above moderate before. The totals are based on all of the units that have received vacancy increases in 2010 (16,633). Page 14
Income Category (% of area median income) Very Low (50-60) Low (80 ) Moderate (100-120) Higher (>120% ) Units with rents in income category rented in 1998 by number and percentage Units with rents in income category rented in 2010 by number and Percentage Difference (-/+) 7973 48% 919 5.5% 7054 5623 34% 1919 11.5% 3704 2532 15% 4285 26% 1753 505 3% 9509 57% 9004 So Costa-Hawkins s effect on the community has been significant. Before its full implementation, 82% of units (48% + 34%) were affordable to those making 80% or less of the area median income, while only 18% were affordable only to those making 100% of the area median income or more. By last year, that had almost perfectly reversed, with 83% of units affordable only to the relatively well off, while a mere 17% were affordable to households earning 80% of median income or less. Page 15