PERSPECTIVES

from the New York State Assembly's
Committee on Ways & Means

Occasional Paper, March 1998, Number 9

Sheldon Silver, Speaker
Herman D. Farrell, Jr., Chair


Trends in the New York State
Correctional System

Growth in New York State’s under-custody inmate population has recently fallen from the eight percent average annual growth experienced between 1980 and 1992, to growth of less than three percent per year between 1992 and 1997. Over the next four years, the rate of population growth is expected to decline even further. The Ways and Means Committee staff projects the State prison population to grow to about 71,300 by the end of the 1998-99 State fiscal year, and to 73,100 by the end of 2001-02. These projections are approximately 500 and 2,900 below the Executive projections for the same periods. In this paper, we examine trends in the State’s criminal justice system with particular focus on the size of the State inmate population.

Between 1980 and 1992, New York under-custody inmate population more than tripled from about 20,000 to almost 62,000, paralleling the national trend.1 In order to accommodate this unprecedented growth, the New York State Department of Correctional Services (DOCS) had to triple the number of prison cells and expand its workforce at nearly the same rate. As a result, correctional services spending grew far faster than overall State spending. Since the 1982-83 State fiscal year, the share of State General Fund spending going toward funding the New York prison system more than doubled (see Figure 1). In the 1996-97 fiscal year the Department of Correctional Services consumed fully 25 percent of the state’s General Fund State Operations budget.

More recently, we find the justice system responding to yet a different trend — the decline in the crime rate experienced since the early 1990’s. Although this is a trend of national scope, it has been acknowledged by experts that New York State, due largely to trends in New York City, is leading the nation in this decline. Following a brief analysis of the decline in the state’s crime rate, we present the results of the Ways and Means Committee staff efforts to project future trends in the inmate population. In addition, we assess the impact the sentencing reforms passed by the State Legislature in June of 1995 will have on the State prison system between now and the year 2002. The objective of these reforms was to hold violent offenders in prison for longer periods of time. Although the sentencing reforms are expected to have a significant out-year population impact, when combined with the impact of a lower crime rate and other reforms, we find only modest growth in the State inmate population by the year 2002.

Population Growth: 1931 - 1972

Between 1931 and 1962, the DOCS population grew from 12,000 to 19,000 inmates, representing an average annual growth rate of 1.8 percent (see Figure 2). Beginning in 1963 the State inmate population actually declined until 1972 when the population hit its lowest point in 40 years. Federal Bureau of Investigation data indicates that incarceration rates were declining nationwide during the 1960’s to early 1970’s.3 New York’s inmate population decline can be ascribed to several factors. First, the Conditional Release program, created in 1965, provided for mandatory early release to parole based on the amount of incident-free time served in prison, better known as accrued “good time”. Second, a 1966 federal court decision known as the Baxtrom decision limited the state’s use of prisons as de facto residences for mentally ill or handicapped offenders who had already completed their sentences.4 Finally, several revisions to the Penal Law were enacted in 1967, including both a redefinition of what constituted a drug offense and the establishment of certain authorized dispositions, such as those for narcotics addicts.5 All of these factors contributed to a decline in the inmate population.

Mandatory Sentencing: the 1970’s

By 1973 the inmate population was on the rise. National incarceration rates also took an upward turn in the mid-1970’s, rising steadily during the 1980’s. In 1973, the New York State Legislature enacted the Rockefeller Drug Laws establishing the minimum drug weights necessary to be convicted of drug sale or possession, and mandating more severe sentencing for certain drug felonies.6 Also in 1973, the State passed the Second Felony Offender Law which imposed both mandatory and extended prison sentences for those convicted of a second felony within 10 years of the first.7 The Violent Felony Offender Law of 1978, designated additional crimes, such as kidnapping, arson, manslaughter, rape, and burglary, as violent felonies warranting longer mandatory prison sentences.8

The Crack Epidemic: the 1980’s

Between 1980 and 1990, the national rate of violent crime increased by 22.7 percent. During the 1980’s efforts in New York City intensified to both enhance law enforcement activities and increase the number of judges assigned to the disposition of felony cases. In 1981, the Office of Court Administration (OCA) increased the number of judges assigned to the disposition of felony cases by 30 percent. During this period, crack-cocaine emerged as a virtual epidemic on the streets of the nation’s cities, including New York City. In 1988, the New York City Police Department established Tactical Narcotics Teams (TNT), which adopted a “buy and bust” strategy in an effort to reduce the volume of narcotics trafficking on City streets. As a result of these actions, New York City arrest rates for drug offenders, which had been rising throughout the period, rose even more sharply (see Figure 3).

The combination of all these factors — tougher drug laws, mandatory sentencing and additional judges, the advent of “crack” cocaine, and finally Tactical Narcotics Team sweeps — caused a dramatic increase in commitments of drug felons to state prisons during the 1980’s. This is best illustrated by the fact that in 1982, 9.5 percent of the inmates in DOCS were under custody on drug charges, while by 1992 that share had grown to almost 35 percent. Meanwhile, the inmate population share for violent felons dropped from 70 percent to 51 percent over the same ten-year period (see Figure 4).

The Decline in Crime: the 1990’s

Across the nation, the crime rate is falling. The estimated number of offenses known to the police fell 4.2 percent between 1990 and 1995. The crime rate per 100,000 inhabitants fell an even more dramatic 9.3 percent during the same period.9 Experts acknowledge that the length of the decline, its magnitude and the number of locations where this has occurred mark a significant downturn. “There is something very real happening. It is not a blip or a passing phenomenon.”10 Nowhere is the decline more dramatic than in New York City, where serious crime is estimated to have fallen 46.1 percent in the six years from 1990 to 1996. The decline in serious and violent crimes being experienced across the nation represents the longest decline in 25 years.

Table 1: Decline in Violent Crime 1990-1996
(Percentage Decline)

New York State New York City
Total Index
(34.0) (46.1)
Violent Crime
(37.8) (43.5)
Property Crime
(33.1) (46.9)
Murder
(49.0) (56.2)
Rape
(22.2) (25.4)
Robbery
(44.9) (50.5)
Aggravated Assault
(29.6) (33.7)
Burglary
(37.3) (48.8)
Larceny
(24.9) (39.5)
MV Theft
(51.9) (59.0)
Source: NYS Division of Criminal Justice Services.
Note: State data includes New York City.

Table 1 illustrates how dramatically violent crime fell throughout New York State and within New York City in particular, between 1990 and 1996.11 Although the largest decline is related to the murder rate, violent crimes of all types have fallen by almost 40 percent. The following trends reported by the Federal Bureau of Investigation indicate that nationally, these trends continued into 1997. Violent crimes committed during the first six months of 1997, including murder, rape assault, and robbery, were reported down five percent from the same period in 1996. Property crimes, including burglary, larceny, and motor vehicle theft, were down four percent during the same period. Although all of the nation’s cities showed a decrease in serious crime during the six-month period, cities of over 250,000 recorded the largest decline of six percent.12

New York City Leads the Nation

The most dramatic decline in the nation has been in New York City where violent crime dropped 43.5 percent between 1990 and 1996. Murders fell 46.9 percent below 1990 figures. Preliminary data for 1997 indicate that murder rates have fallen further, to their lowest level in thirty years. The City recorded less than 1,000 homicides in 1996, for the first time since 1968. Moreover, preliminary data indicate that only 372 murders were recorded during the first six months of 1997, a decline of 28.6 percent from the same period in 1996. Outside of New York City, murders were down 3.8 percent.13 New York City police officials claim that this decline resulted in a larger percentage of solved murders last year. “The simple drop in crime rates, which began in 1990, has helped enormously. Because there were only 984 homicides last year, detectives could devote more time and resources to each killing.”14 The rate of decline we are currently experiencing in violent crime is the sharpest drop since the end of Prohibition. San Francisco, Las Vegas, and several other large cities also report reduced crime rates. Moreover, the suburban metropolitan area, including the surrounding counties of New York, New Jersey, and Connecticut, show a decline in violent crime of 15 percent between 1990 and 1995.

Other cities in New York State have experienced significant declines in violent crime, but the results are more variable. In Schenectady, violent crime in the city fell by 45 percent over the period from 1992 through the first half of 1996. The Onondaga Sheriff’s Department reports that for Syracuse and the surrounding area, serious crimes have declined for the fifth straight year. Buffalo has experienced declines of four percent during each of the past two years.

Experts agree that the decline in the crime rate is very real and very significant. All major violent crimes have declined. Nearly all types and sizes of communities have experienced some drop off, although the most striking changes have been in large cities. Finally, the magnitude of the decline continues to grow with the most dramatic reduction occurring in New York City. After five years, experts agree that the current decline represents a significant shift in the trend.15

Explaining the Decline

A number of theories have been advanced to explain the current decline in the crime rate. While these theories are not comprehensive, they do help us understand the overall trends of the past 15 years. A common element to them all is the dramatic rise of crack cocaine use in the 1980’s.

The Decline of the Crack Epidemic

The decline of the popularity of crack, and cocaine more generally, since the late 1980’s has important implications for criminal activity. Alfred Blumstein emphasizes the important impact of the illegal crack market on the course of criminal activity since 1985. “One explanation for this array of changes from a stable pre-1985 period to a rapidly deteriorating post-1985 period derives from examining the changes in the illegal drug market associated with the introduction of crack cocaine.” The 1991 Survey of State Prison Inmates, conducted by the Bureau of Justice Statistics at the U.S. Department of Justice, indicated that 32 percent of prisoners reported using cocaine or crack regularly, prior to their incarceration; 14 percent reported using cocaine or crack at the time of the offense that led to their imprisonment.16

There are indications that the crack epidemic may have run its course and may have begun to diminish in the early 1990’s. Although abuse of crack-cocaine, and all drugs in general, remains a serious concern for both New York and the nation, there are indications that the intensity of the epidemic, along with the violence associated with it, may be diminishing. New York City arrests related to cocaine for 1996 were down 5.0 percent from 1995 and 16.3 percent from 1990. It is reported that since 1992, declining trends in cocaine emergency department patients are evident in New York City among those between the ages of 18 and 34, and among Blacks and Hispanics.17 Felony arrests related to crack were down by about 1,200 or 6.8 percent between 1994 and 1995 in New York City.18

Demographic changes

Some demographers maintain that with a dwindling population of teenage males and the overall aging of the baby boom generation, a decline in the crime rate was inevitable. New York City demographers report a six to seven percent decline in the City’s population of 15 to 29 year-olds between 1990 and 1995.19 According to Blumstein, one must look at demographic trends in conjunction with the mechanics of the crack market. By looking at age-specific crime rates, Blumstein concludes that individuals began to engage in violent crime at a younger age beginning in about 1985, the time when crack was ascending in popularity in New York City and Los Angeles (and subsequently in other parts of the country). He attributes this increase in juvenile participation in violent crime largely to their recruitment into the crack trade, with its special characteristics. The nature of the crack market - lower prices than other drugs, as well as one-hit-at-a-time buyers - created the need for large numbers of sellers. “Juveniles were the natural source of supply for that labor market.”20

Although age is one of the most important factors affecting crime rates, it is clearly not the only important one. Pointing to the growth in juvenile violence during the late 1980’s, some have warned of the potential for an increasing level of violence as the number of youth between the ages of 15 and 20 increases after the year 2005. The current level of juvenile crime remains a concern for both public and policy makers, alike. However, the above warnings may be unfounded unless, as Blumstein’s analysis suggests, some cultural phenomenon comparable to the emergence of the crack market appears simultaneously.

Fewer Guns

Law enforcement officials in New York City report fewer guns on the street with arrests for gun possession, shootings, and murder with guns all dropping. Again, this may be related to the decline of the crack market. Blumstein observes that, “beginning in 1985, and especially in 1988, there was a steady growth in the use of guns by juveniles, with no corresponding upward trend…in non-gun homicides.” He asserts that, “juveniles, like many other participants in the illicit-drug industry, are likely to carry guns for self-protection, largely because that industry uses guns as an important instrument for dispute resolution.”21

Incarceration Policies

Tougher sentencing laws in many states and localities are seen as contributing to the decline in the crime rate by increasing the incarceration rate and the length of time served. However, the evidence surrounding this claim is mixed. Studies show that from 1980 to 1992, the 10 states where incarceration rates increased the most saw violent crime decrease by eight percent. On the other hand, Florida has the twelfth highest lock-up rate, but ranks first in violent crime. Between 1981 and 1984, the number of individuals incarcerated nationwide rose by 25 percent, while violent crime fell by nine percent. However, the 78 percent increase experienced from 1985 to 1992 was accompanied by an increase in the rate of violent crime of 42 percent.22

Policing Practices

The New York City Police Department (NYPD), as well as some observers, contend that the explanation for the City’s decline in crime lies in a combination of policies that altered the size, approach, and management of the police department. New York City has expanded its police force adding 7,000 officers since 1990, as a result of the “Safe Streets, Safe Cities” legislation adopted by the New York State Legislature at the request of then New York City Mayor David Dinkins.23 Moreover, in its implementation of “Safe Streets,” the NYPD has moved away from the citywide deployment of Tactical Narcotics Teams toward a concentration of its efforts in the City’s most crime-ridden neighborhoods, and toward more community-based strategies commonly referred to as Community Policing.

Consistent with a more community-based strategy, the New York City Police Department has implemented procedures giving precinct commanders greater discretion in developing approaches that work in their neighborhoods. At the same time, commanders are held more responsible for the declines and increases in crime in their precincts. Finally, while precincts now have greater operating flexibility, information on their performance is regularly and rapidly forwarded to central headquarters. Information is reviewed and acted upon quickly. In addition, the department information system, known as Compstat, provides horizontal as well as vertical integration of important crime data across the police, parole, and corrections departments, as well as the district attorney’s office.

An expanded and better-coordinated police presence may be having an impact on the operation of the drug market in a way that reduces the level of violence associated with it. It has been observed that “major drug-selling sites have relocated because of police presence in their area. The police activity has forced them to move some of their drug operation either indoors, to nearby locations, or into areas previously untapped by drug selling.”24 The pressure associated with the awareness of the increased police presence and the associated need to conduct their activities in a more clandestine manner may also be promoting the avoidance of violence as a regular part of doing business.

Other large cities have also seen dramatic declines during the recent period. In Houston, the murder rate declined 48 percent, 46 percent in San Diego, and 45 percent in Dallas. Police departments in these cities undertook similar policies during the same period: expanding the number of police and/or engaging in more aggressive community policing. Other cities in New York State, such as Schenectady and Syracuse, also credit community policing as contributing to reduced crime.

Projecting the Prison Population

The Ways and Means Committee staff projects the New York State Department of Correctional Services inmate population by means of a simulation model.25 The Committee staff criminal justice simulation model is a dynamic, disaggregated flow model comprising two components — a court phase and an incarceration phase.26 The court phase is driven by the total of all violent and non-violent felony arrests. This distinction is useful, given that the sentencing reforms passed in 1995 specifically targeted violent felons. Historically, there has been an average lag of about twelve months between arrest and conviction for those offenders who are ultimately sentenced to State prison. Therefore, the number of arrests provides a strong indication as to the number of prison admissions to expect about one year into the future. The incorporation of arrests into the model permits the analyst to take advantage of this fact.

In order to project prison admissions more than twelve months into the future, it is necessary to forecast both violent and non-violent arrests. The simulation model itself does not project the number of arrests. Future arrests are projected using a multivariate econometric model which combines demographic factors with a statistical method of extrapolating past patterns into the future.27 The model also captures the impacts of the change of administration in New York City, where 70 percent of statewide arrests take place, shifts in policing practices, as well as the seasonality in arrests.28

Figure 3 displays historical arrests going back to 1970, as well as the Ways and Means staff forecast through the year 2002. Non-violent arrests are seen to rise significantly during the 1980’s with increases in the trafficking of crack cocaine and, particularly, with the City’s response in the form of Tactical Narcotics Team sweeps later in the decade. Violent arrests also rose during the late 1980’s, as a result of an increase in the rate of violent crime. However, both types of arrests abated under the Dinkins administration, consistent with the decline in the crime rate which began in 1990 following the implementation of Safe Streets Legislation. This downward trend in arrests was temporarily interrupted upon Mayor Giuliani taking office, but subsequently fell off once again, in line with the reported decline in the City’s crime statistics discussed above. This decline is expected to continue into the out-years, consistent with both demographic trends and the continuing impact of the policing practices which have been credited with reducing crime in New York City and elsewhere.29

The decline in arrests has begun to translate into a decline in the number of admissions to State prison. New court commitment admissions peaked during 1992 at a monthly average of almost 2,100 inmates, and have fallen steadily since (see Figure 5). The Ways and Means Committee staff is projecting a continuation of the decline in the number of new court commitment admissions through 2002, consistent with the projected decline in arrests.

Typically, a new court commitment admission will serve some portion of his or her maximum sentence, and be released to supervision by the New York State Division of Parole. The duration of time which an offender spends in prison depends on two factors: the sentence imposed by the court and the decision by the Parole Board as to how much of the sentence will be served. A sentencing framework set in statute establishes parameters which judges must abide by when imposing sentence on a convicted offender. Portions of this framework were reformed in 1995 to keep violent offenders incarcerated for longer periods. However, the reform of sentencing guidelines is but one way the State can ensure that violent offenders serve more time in prison. The Parole Board can alter its release policies by reducing the rate at which it grants parole to offenders at their first hearing. Since the 1992-93 State fiscal year, the Parole Board has reduced this rate from 64 percent to about 55 percent. The Committee staff incorporates this shift in Parole Board policy into its baseline forecast.

Some portion of the individuals released to parole supervision will return to prison as parole or conditional release violators. A relatively small number of inmates are released from Department of Correctional Services custody for reasons other than the completion of their sentence, such as admission to mental hygiene facilities or for court appearances. Almost all of these individuals will return to custody. The Committee staff projects future parole and conditional release violators, as well as re-admissions of individuals released for these other reasons.30 The parole and conditional release violator model incorporates the number of new court commitments who have completed their initial terms of incarceration, the volume of arrest activity (as an indicator of the intensity of police activity), as well as factors which capture past trends and administrative policy changes. The “other” admission model incorporates lagged values of new court commitment admissions, as well as factors which capture past trends. The Committee staff baseline forecast projects a decline in the numbers of admissions of both parole violators and other admission types in the out-years, consistent with lower release rates.

 

Sentencing Reforms of 1995

With the passage of the 1995-96 State budget, the New York State Legislature enacted a sweeping package of reforms which altered the way judges may sentence violent felony offenders.31 The reforms identify four classes of violent offenders: first-time felony offenders, second-time felony/first-time violent felony offenders, second-time felony/second-time violent felony offenders, and persistent violent felony offenders (those having three or more violent felony convictions). The Committee staff projects the impact of the sentencing guideline reforms on each of these groups using the computer simulation model described above.

For first-time felony offenders, the new reforms parallel the Second Felony Offender Law enacted in 1973 by mandating that offenders receive minimum sentences which are one half the maximum sentence, rather than the one third, as was the previous practice for first-time felons of all types. The reforms mandate determinate sentencing for both second-time felony/first-time violent felony offenders and second-time felony/second-time violent felony offenders. Offenders must be given a single sentence within a specified range and must serve at least six-sevenths of that sentence before their first Parole Board hearing. Finally, persistent violent offenders must be given lengthier minimum sentences, their maximum sentences already fixed at life-sentences under prior guidelines.

The Sentencing Reform Act of 1995 introduced determinate sentencing to the State’s criminal justice system. Therefore, there was very limited experience to draw from for the purpose of impact analysis. At the time the projection was being constructed, over 3,000 cases had been adjudicated as second felony offenders and given determinate sentences under the new law. It was assumed that on average, those given determinate sentences would spend six-sevenths of that sentence in prison.32

Both first-time violent felons and persistent violent felons will continue to receive indeterminate sentences as before. For first-time violent felons, it was assumed that judges would mete out maximum sentences with roughly the same frequency as before. However, the frequency of minimum sentences would change to reflect the new law. Before the new law went into effect, approximately 35 percent of persistent violent felons received minimum sentences of ten years or more. All of these offenders will receive sentences of ten years or more under the new law, significantly increasing their average sentence length.

Inmate Population Projections:
1997-98 through 2001-02

The results of the Committee staff simulation appear in Table 2. Absent the 1995, 1996, and 1997 reforms, the State prison population would be expected to experience modest growth of 0.7 percent for the 1998-99 fiscal year, followed by declines throughout the forecast period. After peaking near 71,000 at the end of the 1998-99 fiscal year, the inmate population would decline by more than 1,400 individuals by the end of the 2001-02 fiscal year, due to a continuing decline in the crime rate and the resulting decline in arrests. The decline in crime, and the attendant decline in arrests, produces fewer cases to be adjudicated and, therefore, fewer convictions. By March 2002, the inmate population would have been projected to total 69,507 under prior law.

The effect of the sentencing reforms enacted in 1995 and assault reforms enacted in 1996 is to increase the number of people who are incarcerated, with the impact growing in the out-years as the system begins to increasingly feel the impact of the longer sentences given to violent offenders. In contrast, merit time reform enacted in 1997 will have the opposite effect. By the end of the 2001-02 State fiscal year, the reforms are expected to result in the retention of over 3,600 inmates relative to the projected baseline population. The Committee staff projects modest growth in the State’s inmate population through March 31, 2002.33

Table 2: State Prison Inmate Population Projections

Baseline
Reform Impact Total
End of Fiscal Year
Population
Percent
Growth
Population
Impact
Population
Growth
Percent
Growth
1997-98 70,437 -- 310 70,747 -- --
1998-99 70,919 0.7 375 71,294 547 0.8
1999-00 70,560 (0.5) 1,469 72,029 735 1.0
2000-01 70,014 (0.8) 2,576 72,590 561 0.8
2001-02 69,507 (0.7) 3,637 73,144 554 0.8
Source: Ways and Means Committee staff estimates.
Note: The projected total population is equal to the projected baseline population plus the projected impact of the Sentencing Reform Act of 1995, assault reform enacted in 1996 and merit-time reform enacted in 1997.

Conclusion

New York continues to spend an ever larger portion of its budget on the operation of State prisons. In the past fourteen years, the New York State Department of Correctional Services’ share of General Fund State Operations spending has grown from 9.5 percent to 25 percent. Historically, the prison system has been almost entirely financed through General Fund revenues. Although the Department has received increased financial assistance from the federal government, as well as other “special revenue” sources, such as the sale of prison-made goods and charges to inmates for both commissary purchases and collect telephone calls, 90 percent of the Department’s operating budget is still supported by tax dollars from the General Fund. However, the current trend toward slower inmate population growth provides an opportunity to focus on alternative uses of State revenues, such as crime reduction strategies. A recent study found that strategies designed to prevent crime, along with conventional law enforcement methods, constitute the most cost-effective method for confronting what many of our State’s citizens deem as one of the most serious problems they face — crime.34

In State Fiscal Year 1997-98 the Legislature authorized a two-year plan for the construction of an additional 3,100 beds.35 This plan includes the construction of one, 750 double cell maximum security prison, with beds coming on-line in May 1999 and completed in September of that same year. The plan also calls for the construction of eight, 100 double celled, special housing units located at existing medium facilities. Construction for these smaller facilities has already begun, with the first building at Greene Correctional Facility due to open in April 1998. The completion of Cayuga, Lakeview and Orleans Correctional Facilities will immediately follow in May 1998. The next phase of construction at Midstate, Fishkill and Collins Correctional Facilities is due to come on-line in June 1998, while the final phase is scheduled to be completed at Gouverneur Correctional Facility in August 1998.

In an additional effort to assist the Department of Correctional Services in managing the prison population, the Legislature authorized the establishment of a new Local Jail Utilization program as part of the 1997-98 budget agreement.36 This new program makes availabile up to 1,800 additional beds through the rental of vacant local jail beds for which the State will pay actual operating costs, up to $100 per day.

Since the two year construction plan was authorized in 1997-98, the major population indicators have continued to drop. The Committee staff estimates that between now and the end of the 2001-02 State fiscal year, the prison population will grow by an additional 2,400 inmates. With the decline in the crime rate, the authorization for 3,100 new beds, and the 1,800 additional local bed capacity, the Department of Correctional Services now has sufficient capacity to manage current and future population needs.

Prepared by the NYS Assembly Ways and Means Staff.


Other Available Publications

Statistical and Narrative Summary of the Executive Budget, Fiscal Year April 1, 1998 to March 31, 1999, January 1998.

New York State Economic Report 1997 & 1998, March 1998.

New York State Revenue Report 1997-98 & 1998-99, March 1998.

“Trends in Public Assistance Spending in New York State”, March 1998.

“Trends in Medicaid Spending”, March 1998.

For more information on publications of the Assembly Ways and Means Committee, please contact:
Deborah Priest, Director of Information Center,
Empire State Plaza, Agency Bldg. 4, 14th Floor
Albany, New York 12248, (518) 455-4780

New York State Assembly
[ Welcome Page ] [ Reports ]