Skip to content

Latest commit

 

History

History
1072 lines (830 loc) · 84.2 KB

report.md

File metadata and controls

1072 lines (830 loc) · 84.2 KB
Raw

Why All the Ruckus?

Table of Contents

Executive Summary

Overview

This is an initial analysis of New York City Department of Buildings data performed by Abe Lerman and Xiaofeng Cao of NewTrails Data Science.

The date of publication of our report is March 2nd, 2020 - our research was conducted over several preceding months, primarily during the fourth calendar quarter of 2019. The executive summary will provide some context on various aspects of the project, each of which - and more - will be discussed in greater detail throughout the remainder of this report. For more information, you can visit the GitHub repo for this project, where you can:

  • download our code and analysis notebooks
  • contribute directly to the materials
  • notify us of any issues

You can also reach out to us over email.

Problem Statement

Facing an onslaught of construction-related, quality-of-life issues and earning little in the way of respite through filing complaints with the city, we set out to study building and construction trends in NYC. Our hope was to gain a better understanding of the NYC real-estate construction landscape. Specifically, we sought to answer the following questions:

  1. is NYC currently over-indexing, compared to historical trends, with respect to construction activity?
  2. is the construction that's going on of a more harmful nature, and / or less respectful to residents' quality of life?
  3. is the city enforcing building-code adherence with the same level of rigor as it has historically?

Methodology

In trying to answer these questions, we ingested several datasets publicly released by the New York City Department of Buildings (DoB) through the OpenData platform and performed our own analysis on this data. We call this an initial analysis because there's plenty more to do: in this report, we take a high-level look at certain corners of the available data, focusing only slightly on other corners and ignoring some others altogether. A key constraint to note is that we looked exclusively at NYC-level data - i.e. we did not drill down to the level of a neighborhood, or even a borough. Another is that, unless otherwise noted, we focused our analysis on the years 2000 through 2019.

Analysis Findings

1: An Increase in Construction Activity?

Regarding recent construction activity as compared to historical trends, NYC saw a tremendous run-up in applications and permits in the post-recession years, with 2016 and 2017 exceeding* any of their respective prior yearly totals. As is described below in Application Filings, we could not really study permit applications beyond 2016. But, we were able, to some extent, to study permit issuances, for which 2018 was a peak. "Construction equipment" was, far and away, the most-issued permit type* over the time-frame we studied.

We reiterate that we are painting in broad strokes - construction activity almost-assuredly varies throughout different boroughs and neighborhoods, both in terms of the raw number of construction projects, as well as in the magnitude of such projects. That said, it would be correct to say that New York City, on the aggregate, has seen unprecedented levels of permit issuance in the last few years. With that in mind, we have likely (in one way or another) been feeling the effects of of such issuance activity in recent years.

2: An Increase in the Harmfulness of Construction Activity?

As for the second question - both complaints and violations are at all-time highs*. This could very well be a result of the fact that construction activity in general is at record levels. However, it should be noted that both complaints and violations, proportionate to the number of permits issued, have been on the rise*. This annually tabulated complaints-to-permits ratio bottomed out at 56% in 2015 but has since risen, hitting 74% in 2019. Ditto for the similarly tabulated violations-to-permits ratio, which after hovering around 28% from 2013 - 2015 and has since also risen, coming in at 48% last year. Neither of these KPIs have reached their records highs (posted in 2008 and 2009, respectively) but the increases each has seen in the past 4 years are, nonetheless, large. You can find more on this below. Additionally, construction-related violations have sky-rocketed over the last several years, more than doubling since 2014* - there were 71.5k construction-related, DoB-issued ECB violations issued last year. For reference, there were 5.4k elevator-related DoB-issued ECB violations issued in 2019, and an even-smaller 1.5k of these violations tagged with quality-of-life.

It's hard to answer this question definitively, given the potential for confounding. Has it gotten easier, at least in the last 10 years, to file a complaint against a landlord or nearby construction project? The city has certainly made updates to its 311 program. Is it possible that constituents have become less tolerant of construction, even if the relative harmfulness of that construction hasn't really gone anywhere? Given the increases in rental costs, along with the increases in construction-related activity, yes. And, to reiterate, we took a very high-level view in this analysis - the real answer to this question would probably vary over different areas of the city. But with all that said, there do seem to be indications that construction activity has in recent years taken a turn that is more harmful to NYC residents.

3: Is the City Rigorously Enforcing the Building Code?

Finally, is the city rigorously enforcing the building code? The answer is complicated. After refusing to really climb for the greater part of the last decade, the median annual ECB penalty issued (explained in more detail below) did rise in 2019*. (This rise, by the way, holds true when grouping by the year in which the penalty was issued, not by when the hearing was held - the latter of these methodologies yields results that lag those the former.) This could obviously mean a number of things; for instance, that:

  • the city is trying to make more severe the consequences for violating important aspects of the building code
  • there are just more, severe violations of the building code
  • more people are being caught for violating the building code
  • this rise in median annual penalty was a statistical blip, and we will eventually revert to the mean (or median)

The truth may very well lie in some combination of these factors.

The allocation of responsibility for the issuance, assessment, and collection of violations and penalties is not entirely straightforward. The DoB handles the issuance of violations, which can be either "DoB" or "ECB" (Environmental Control Board) violations. The DoB's own violations tend to be more administrative in nature, while ECB violations that the DoB may issue tend to cover the more severe and "quality-of-life" impacting issues, as per our own research and what we were told by folks we spoke with in the NYC government. The ECB is a subdivision of the Office of Administrative Trials and Hearings, which operates laterally to other agencies (like the DoB) who rely on it for adjudication. Depending on the severity of the ECB violation, a respondent may be able to avoid a hearing and / or reduce the ultimate penalty imposed at that hearing, through either correcting the issues that led to the violation, pre-paying by mail, or agreeing to some other set of stipulations. If none of that happens, the ECB administers a hearing on the violation, which may result in the imposition of a penalty. If a penalty is imposed it would likely be due immediately, unless otherwise stipulated, and the ECB would be responsible for collecting said penalty*.

With all of that out of the way, we can note the following:

  • the median annual paid ECB penalty, even when grouping by hearing date, has spent the last 3 years at $0*
  • as the number of issued ECB violations has sky-rocketed over the last few years, the annual ratio of paid-to-imposed penalties has plummeted*
  • despite starting the previous decade at around 25%, the ratio of the total $-value for in-default (no-show at hearing with an outstanding balance) ECB violations, proportionate to the total $-value of all imposed ECB violations, has held at over 50% for several years*
  • the median difference in days between dates of hearing and issuance for DoB-issued ECB violations last year breached 100 for the first time (in our data, since 2000) and is up 88% from 2013 *
  • the temporality with which the DoB issues its own, administrative violations, which may be a part of the procedure for enforcing penalty payment, is surprising in its irregularity*
  • the balance of unpaid DoB-issued ECB violations last year measured out at 42% of the DoB's total annual expenses*

More analysis on DoB-issued ECB penalties can be found below, but suffice it to say that the ECB looks to be overwhelmed by the increase in issued violations over the last few years, and that it may be having a hard time collecting the penalties for those violations.

In a relevant argument, Danielle Chaim recently noted in Columbia Law School's CLS Blue Sky Blog that a large number of firms all committing tax avoidance at the same increases the likelihood of the government becoming overwhelmed in its enforcement, thus decreasing the probability of enforcement action. It is important that the city engage in, and have the proper resources to engage in, rigorous enforcement of its building code, lest developers, landlords, and others come to believe that there are only light consequences for flouting the building code. Even those industry practitioners who hold themselves to a higher standard may face pressure to cut corners if their competitors are repeatedly able to get away with delivering faster or cheaper solutions through kicking the metaphorical "consequence" can down the road. As we've already seen, such a culture can produce severe consequences.

Recommendations - Summary

The city needs to be more systematic in its enforcement of the building code, which includes ensuring that:

  • penalty payment is enforced
  • ECB hearings are held in a timely manner
  • penalty values grow at a reasonable rate

Next Steps - Summary

We could further extend this analysis to borough-level and neighborhood-level. Moreover, roping in additional, economic variables could provide additional context. Finally, linking applications and permits to complaints and violations could help the city keep a handle on bad actors and preempt the types of work that are at a high-risk of heading in the wrong direction.

^

Background

To offer a brief history of NewTrails, we were founded last year in the hope of being able to offer flexible but robust data engineering and machine learning services to a variety of different types of organizations. The work last year was carried out primarily by Abe but we are now a team of three. Xiaofeng - who started working on this report as an intern - has joined as a Data Scientist, and Andriy Noble has joined as our Principal Security Engineer, enabling us to enhance our offering with a suite of cyber-security-oriented data science solutions. We also maintain an, if you will, distributed network of on-demand talent to help out as necessary, and we plan on adding additional, full-time capacity this year.

The idea for this project came about towards the end of last summer - we were brainstorming on what Xiaofeng's next project should be, as she had just successfully completed an audit of some of our analysis programs, and she was looking for a real-world, gnarly, data wrangling problem. We stumbled into what would become the underlying idea for this report by means of having our conversation interrupted by some incredible construction noise right outside the window.

Abe had spent months dealing with a variety of problems at his apartment. First, there was the late-night construction across the street. Finding little in return for the 311 complaints he'd made, he reached out to his councilwoman's office, who was responsive and empathetic, but ultimately just encouraged him to continue calling 311. This construction went on for some time and was severe enough that it, along with his and his neighbor's complaints, was all picked up by the New York Times. There was also the construction for the restaurant downstairs, some of which was taking place - illegally - on the weekends. And, of course, there were the myriad maintenance problems inside the apartment itself. As all of this continued to happen, right in the heart of the Flatiron district, we wondered whether this was a localized suite of issues, or a representation of a larger problem?

It had long been a hope that NewTrails would occasionally carve out time for interesting open-source work. Plus, what better way to gain familiarity with the new (to us) and exciting industry of real estate construction? And so it was that we set out to put our data science abilities to use in gaining an understanding of NYC's building code and the DoB's enforcement of them. An early theory was that key, problematic actors were working as force multipliers in the construction space, engaging in bad behavior that would allow them to get through certain work-flows at higher speeds and / or lower costs, thus encouraging competitors to degrade their standards lest they lose market share. Given the level of depth and data-access required in order to properly solve such a problem, and our correspondingly limited time and lack of access to any non-public data, we whittled our ambitions down to answering the three, core questions laid out in our problem statement over the course of a few months of part-time work.

^

Data Overview

Application Filings

  • data
  • assumed primary key: job
  • assumed primary date column: pre_filing_date

This dataset corresponds to the applications filed, applications which - if approved - would yield permit(s), generally one permit for each type of work performed (e.g. electrical, plumbing, etc.). Unfortunately, this dataset was unusable for us post-2017 (and even somewhat for 2017) due to the fact that it does not include applications submitted through the DoB's new platform, and cannot be deterministically linked to those new jobs, at least not in any sort of way that would have fit with the level of granularity we have tried to achieve in this report.

The DoB has released a new system for permit application and issuance that it calls DOB NOW. This system was used for about 1% of permit issuances in 2017, a majority of permit issuances in 2018, and almost all permit issuances last year. The data (which we've linked to in the above paragraph) corresponding to the applications filed through this system, at the time of this writing, included neither a unique application identifier that linked to previous system, nor any sort of date field. After making multiple requests for a date field, we were told that a ticket had been filed, and that there was no time estimate corresponding to when this field would be added.

See the DOB NOW appplications data here.

Permit Issuance - Old

  • data
  • assumed primary key: combination of permit_si_no and
  • assumed primary date column: issuance_date

This dataset corresponds to permits issued by the Department of Buildings since 1989.

Permit Issuance - New

  • data
  • assumed primary key: job_filing_number
  • assumed primary date column: issued_date

Provides all permits approved through DOB NOW (first put to use in 2016). We were able to link this data to that of the old system, at least at the level of a year. However, there are differences in the permit types found in each system, which limited what we felt able to extract from the permit-type information.

Complaints Received

  • data
  • assumed primary key: complaint_number
  • assumed primary date column: date_entered
  • de-duplicated on complaint_number due to the reappearance of complaints in the data, as complaint status is updated

This dataset corresponds to complaints received by Department of Buildings (DoB). It includes complaints originating with 311, those that were called into the DoB's Customer Service Center, and those that were directly entered into the system by DoB staff (e.g. building inspectors).

ECB Violations

  • data
  • assumed primary key: isn_dob_bis_extract
  • assumed primary date column: issue_date
  • de-null issue_date

The DoB issued ECB/OATH violations when a property is not in compliance with either New York City construction codes or zoning resolutions*. These violations are adjudicated by OATH and generally apply to offenses that are more harmful in nature than those covered by the more-administrative, non-ECB DoB violations. This dataset includes details of ECB violations going back to the 1990s. We dropped violations with null issue_date values, as this date field was core to our analysis and these values accounted for less than 0.005% of all values issue_date.

Please note that, unless we explicitly state otherwise, all "violations" discussed are taken to be DoB-issued ECB violations.

Also, please note that the DoB seems to update both paid- and imposed-penalty data as new information comes in. As such, the total figures (as well as the related ratios) pertaining to paid and imposed penalty values may come out differently, depending on when this data is pulled. Our charts are based off of data that was current as of mid-February 2020.

DoB Violations

  • data
  • assumed primary date column: issue_date
  • de-null issue_date

This dataset provides information on the DoB's own violations, which generally correspond to violations that are more administrative and less harmful than those covered by ECB violations. Some of these violations, e.g. those corresponding to make the proper Boiler or Elevator filing with DoB, might be periodically auto-generated, which explains some of the irregular temporality we saw in this data.

^

High-Level Trends

Figure 1

As many in real estate already know, there's a cyclical nature to construction activity. This is common in high-capEx industries such as construction and oil extraction where getting to a revenue-producing product requires heavy upfront spending. This leaves market participants sensitive to long-term signals, due to a heavy exposure and lack of agility. These signals, however, may be both imperfect and followed by many a market participant. Crowding can lead to supply gluts (price dips), pullbacks (price increases), and eventual reinvestment (in pursuit of the higher prices). We saw large increases in permit applications and issuances in the years immediately preceding the Great Recession (mid-aughts), and we saw another large run up from 2009 to around 2016 or 2017.

We also saw that, on the aggregate, complaints and violations track each other pretty well. However, we saw that these two (complaints and violations) seem to lag behind permit applications and issuances by about two years. We look to be in the midst of a plateauing, or maybe even the beginnings of a downswing, in permitting activity. It's worth mentioning that there are more permits than applications each year since a single application might cover more than one type of work at a given job, with each type potentially receiving its own permit from the DoB.

As has already been noted the DoB has implemented a new system for applications and permitting known as DOB NOW. This is the Department of Building's self-service online tool that enables applicants to do all business with the DoB online. We included the DOB NOW permits starting from 2017 (the number of permits issued through DOB NOW in 2016 was negligible).

However, as has already been mentioned, we were limited in how we could make use of the DOB NOW applications data, due to there (at the time of this writing) not being a date column in that data set. As such, we estimated total annual applications from 2017 to 2019 based on applying the applications-to-permits ratio in 2016 to the number of permits issued in each of these subsequent years. This method was chosen for the sake of simplicity. The estimated numbers are showed with a dotted line.

Figure 2

Autocorrelation represents "the correlation of a signal with a delayed copy of itself," as per Wikipedia. Figure 2 shows the autocorrelation in monthly percent changes in permits, complaints, and violations. As with correlation, an autocorrelation of 1 indicates a perfect and positive correlation, while an autocorrelation of -1 shows a perfect and negative correlation - an autocorrelation of 0 indicates a lack of correlation. We looked at the autocorrelations of lags / shifted-values 1 through 30.

The permits data exhibited very strong autocorrelation at a lag of 12, i.e. between permits issued in the current month and in the same month in the previous year, reaching .71. We imagine that this is driven by auto-renewals. Autocorrelation in complaints also reached its highest point, 0.45, at lag 12 - this could be driven by auto-generated complaints, but we don't know. In violations, we again found the strongest autocorrelation, 0.23 at lag 12.

^

Applications and Permits

Figure 3

New York City issued 197,556, 195,817 and 192,399 permits each year from 2017 to 2019, including the regular / old-system DoB permits as well as those issued by DOB NOW.

To break it down by system (over the same, three-year period):

  • 195,343
  • 169,377
  • 152,212

permits were issued through the old system, while:

  • 2,213
  • 26,440
  • 40,187

went through DOB NOW. The DoB has been pushing its DOB NOW system as the desired standard and the system has been accounting for an increasing proportion of permit issuances each year.

Applications and permits tracked each other well over the time period study - the vast majority of applications are approved. In 2017, there are 94% more permits issued, as compared to the year 2000. The climb in permit issuance since the recession shows that the real estate market has been recovering well, but also that it may be cooling off since hitting its all-time high in 2017

Figure 4

We can see here the tremendous growth in Construction Equipment permits issued, although it would be nice to know more about what this means - do these types of permits tie more closely to new-building construction?

It's important to note that, to thoroughly consider whether the city has over-indexed on construction, one should explore applications and permits in the context of additional variables. For instance, what is the relationship between permitting rates and population density, in different parts of the city? Manhattan is geographically constrained by water on all four sides. So, outside of maybe the Hudson Yards project, where a not-so-inhabited nook of the city was developed freshly in new ways, construction will be felt much more heavily in Manhattan than, say, in deep Brooklyn.

One of the great catch-22's of modern life in NYC can be found in the interplay between cost of living and construction-related, quality-of-life issues: we want reasonable rent, but we hate the effects of the construction that helps increase housing supply to the point where prices may fall.

^

Complaints and Violations

Figure 5

The above ratios help to put the number of annual complaints and violations in context - i.e. let's scale each measure down by permit issuance, our proxy for construction activity.

Complaints-to-permits reached 74% at the end of 2019, meaning that, for every 100 permits issued in 2019 there were 74 complaints made. To put this figure in context, this ratio sat at 43% in 2000, 95% in 2008, and 56% in 2015. One takeaway might be that real estate booms coincide with some combination of bad behavior, frustrated residents. That said, maybe the the 2008 peak had something to do with a more streamlined complaint-filing process - more work could be done to flesh this out. It's also worth highlighting the growth this measure has exhibited over the last few years - it's almost 50% higher than it was 4 years ago.

Violations-to-complaints came in at 48% last year. This measure was at 40% in 2000, 58% in 2009 (record high), and 28% in 2015. Again, opportunities for confounding abound - but the growth since 2015 is real.

Figure 6

In this case, we filtered the data to 2006 - 2019 so that we could stick to plotting just 3 complaint types.

The top complaint type, Illegal Conversion, is defined by the city as "an alteration or modification of an existing building to create an additional housing unit without first obtaining approval from the New York City Department of Buildings," which have declined a bit over the past few years.

Working without a permit came in first for several years in the period studied, and placed second in other years. What type of work this corresponded to, we don't know, but it makes a person wonder what kinds of maintenance and construction activities are going on outside the purview of the authorities.

The number 3, and maybe least-assuring complaint type on the list, is that of Elevator. What we've plotted here is actually the result of rolling two, different complaint types up to a single label. These complaint types existed at different times, with the previous elevator category, Elevator-Danger Condition/Shaft Open/Unguarded in recent years being substituted out in favor of two new categories, with one being more severe than the other. What you see here is the old category through 2017, and the new, more-severe category for 2018 - 2019. See the chart's corresponding data for more information.

The official, somewhat-complete complaint-type-code mappings can be seen here.

Figure 7

We can see here that Construction violations have absolutely skyrocketed since 2013, increasing from just-under 25k in that year to more than 70k in 2019. That doesn't mean there's no ambiguity - might the Construction category simply be absorbing some of what was previously considered Unknown? Could the Unknown category have absorbed some of what was previously tagged with Quality of Life? It's hard to know. Regardless, the increase in Construction violations is large enough to stand on its own, offering some support for the "yes, construction has been getting more harmful" argument.

^

Violation Penalties

Figure 8

To better understand the underlying penalty data, we grouped the data by the years corresponding to two different date fields (hearing date and issue date) and calculated both the mean and median for each year. Note that issue date is the date on which a violation is issued, and hearing date is the date of the most recent, scheduled ECB hearing. A hearing presents an opportunity for the respondent named on a violation to either admit to or contest it.

In statistics, the mean and median are two main measures for the central tendency of a dataset - i.e. where do the values of an at-least-somewhat random variable tend to fall? Or, more colloquially: if you had to expect one value from a dataset, what would it be? Expectation refers to a data's mean, but the mean can be unduly influenced by outliers, which the median is much more robust to. Thus, the measures are related but different, and their comparison can prove valuable.

A positive (negative) difference between mean and median would indicate that the data is positively (negatively) skewed - i.e. there are is a long / dispersed "tail" stretching right-ward (left-ward) away from the probabilistic center of the data. In other words, skew indicates asymmetry - when the data falls away from its center, it tends to fall farther in one direction than in the other. The large, positive difference we can see between mean in median in this chart indicates that a relatively small number of high-valued penalties are "pulling" the mean away from the center of the data. Thus, the expectation for an imposed penalty is much higher than the actual midpoint of all imposed penalties.

Surprisingly, at the end of 2018 to 2019, both the mean and median measures (in the case of the median, only by issue date) sharply increased, indicating that the city issued many expensive DoB ECB violations.

Figure 9

In this chart, we look at what's actually been paid by respondents, as opposed to what's been imposed by the city. Some of the trends that can be seen in the above, imposed chart hold true here, too. But, the mean median measures of paid penalties have tumbled since the end of 2018, which when compared to the above data on imposed penalties indicates that many ECB penalties issued in the last two years have not been paid.

To some extent, this makes sense - there may very well be some kind of a natural gap between when penalties are paid and when they're imposed. But, when the median paid penalty spends multiple years at zero - meaning that the paid amount for at least half of all penalties imposed or heard (this trend was similar for both date fields) was zero dollars - during periods of time when the median imposed penalty was more than $500, and when the mean imposed penalty was more than $1000, that seems suspect.

Figure 10

Taking the insights described under Figure 9 further, we can examine the paid-to-imposed ratio, grouped annually by hearing date. This ratio measures the total dollar value of penalties paid by respondents as a proportion of the total dollar value of penalties imposed by the ECB. The paid-to-imposed ratio went from 48% in 2017, to 37% in 2018, to 23% last year. Why the steep drop?

The blue bar shows the number of DoB-issued ECB violations issued. As we can see, after declining in the years immediately proceeding the recession, the number of violations issued has broken records each of the last 3 years, standing at over 90k for last year.

With that said, 23% does seem quite low, no?

What "good" or "bad" would mean here could surely depend on a lot of things, so let's lay it out over a couple of points. A degradation in this ratio means, unambiguously, that violations are being issued at a rate that is higher than the rate at which penalties are being paid. To gain a sense of the magnitude or severity of the problem, we can consider the rate at which the ratio is degrading to the rate at which the number of issued violations is growing. With that, paid-to-imposed fell by 22.12% between 2017 and 2018, and the number of issued violations grew by 12% over the same period - those are very different change rates! Thus it might suffice to say that the city's resources, when it comes to collecting on the penalties it's imposed, are stretched pretty thin.

Figure 11

To further analyze paid-to-imposed ratios, we grouped the DoB-issued ECB violations by hearing status and hearing year. Normalized penalty imposed represents the proportion of total annual penalty imposed that each type of hearing status accounted for. We looked in detail at two hearing stati, DEFAULT and IN VIOLATION, because these stati together accounted for almost all of the total dollar value of imposed penalties last year.

Violation Status: DEFAULT

A hearing status of DEFAULT indicates that the respondent failed to appear for scheduled hearing and did not pay the summons prior to the hearing. ECB may add additional penalties on top the existing one(s).

We can see that violations with a status of DEFAULT have each year, since 2012, accounted for about 50-60% of the total dollar-value of all DoB-issued ECB violations, while the paid-to-imposed ratio over this period of time has typically stayed below 20%. Moreover, (as seen in the text accompanying the next figure) given that it usually takes less than a year for the city to hold an ECB hearing for a given violation, the 2018 paid-to-imposed ratio of 7% for DEFAULT violations (which accounted for 52% of the total dollar-value of DoB-issued ECB violations that year) seems outrageous - that's a lot of money that should be going to the city.

Without severe violations in place for defaulting on penalty payment, these penalties could easily come to be seen as simply a cost of doing business, payable whenever it is convenient to do so - who's to say that this isn't already the case? Stronger default penalties could make a real impact on the timeliness with which these violations are paid off.

Additionally, with such a high % of penalty dollar-value in default each year, is there any way that the city is tracking all of these down? An overburdened system presents opportunities for bad actors to slip through the cracks. With that, making an impact on payment timeliness probably means making an impact both to the city's bottom line, and to the behavior of the various players in its ecosystem.

Violation Status: IN VIOLATION

DoB-issued ECB violations with hearing status IN VIOLATION in the last decade typically accounted for around around 40% of the total dollar value of all DoB ECB penalties imposed each year. The paid-to-imposed ratio for these penalties peaked at 76% in 2016 and has since fallen - to 75% in 2017, 67% in 2018, and 48% last year. One reason for this decline could be the massive increase in DoB ECB Violations issued over the last few years.

Violation Status: All

There are nine types of hearing status as following:

  • IN VIOLATION: Definition unknown.
  • CURED/IN-VIO: Enforcement agency notes indicate that the violating conditions have been corrected prior to hearing.
  • POP/IN-VIO: Definition unknown.
  • STIPULATION/IN-VIO: Admission of guilt – extends compliance time by an additional 75 days from the first scheduled hearing date.
  • PENDING: Definition unknown. Penalties of this status couldn't be found for most of the last decade, but exhibited a strangely high paid-to-imposed-ratio of 66% in 2017 and accounted for 0.15% of all DoB ECB penalties imposed that year.
  • DEFAULT: Respondent failed to appear for scheduled hearing and did not pay the summons prior to the hearing.
  • WRITTEN OFF: Summons has been written off and nothing is due.
  • DEFAULT: Respondent failed to appear for scheduled hearing and did not pay the summons prior to the hearing.
  • DISMISSED: Dismissal – No Penalty Imposed; If you prevail in contesting your violation, you will not owe any penalties and your violation will be dismissed. However, the Department may re-inspect, reissue a violation, or appeal the decision.

Figure 12

All ECB violations need to be adjudicated by the Office of Administrative Trials and Hearings (OATH), so we analyzed the day-differences between hearing and issue dates across all DoB-issued ECB violations, broken out by year.

The above chart shows us the mean and median number of days that it took for OATH/ECB to hold a hearing (i.e. the delta, in days, between hearing and issue dates) from 2000 to 2019 (by hearing date year). As we have mentioned in figure 8, both mean and median measure the central tendency of a distribution, and a positive difference between the two indicates positive skewness, which we can see here. This means that there are likely extreme values that are "pulling" the mean higher.

Both the mean and median days-delta reached record-high values in 2010. For all the hearings that year, violations had been issued, on average, 195 days earlier; the median difference was 87 days. The fact that these measures were greater than they were in any previous years is likely a result of the sharp increase in ECB violations (shown in figure 10) that happened between 2007 and 2009. In other words, the city was probably overwhelmed.

The median is much more robust to outliers than the mean is, so an up-tick in the media days-delta would be a solid indicator of a slowdown in the ECB adjudication process. With that in mind, take note of the fact that between 2015 - 2019, the median days-delta value has grown by 87%. It is therefore likely reasonable to assume that the city is, once again, overwhelmed.

Figure 13

Quantiles (%) 2017 (days) 2018 (days) 2019 (days)
0.00 12 1 1
5.00 45 46 49
10.00 47 48 53
25.00 49 54 69
50.00 68 86 105
75.00 133 158 175
90.00 212 246 253
95.00 272 311 320
99.00 697 668 634
99.90 3092 3026 2891
99.99 6279 7447 3880
100.00 10347 9639 6506

We further analyze the days-delta values described above through looking at some different percentiles for this measure, in 2017, 2018, and 2019 (by hearing date year). The value for a given percentile represents the value below which a given percent of the data lies. For instance, the median (it would be reasonable to say) is the 50th percentile.

We highlight the 95th percentile to substantiate the claim that most violations have been heard within a year. In 2017, 95% of DoB-issued ECB violations that had a hearing that year had done so within 272 days of issuance. For 2018, this number was 311 days. A penalty, if imposed, is imposed at hearing and is (as far as we know) due pretty much immediately, unless some other agreement has been made.

Thus, a full year (or 365 days) seems a good window over which to judge penalty collection, which adds strength to our paid-penalty analyses accompanying Figures 9 and 10.

Finally, it is worth mentioning that we have about 30 years of DoB ECB violation data, so it is not inconceivable that, over this entire time period, a small number of violations managed to ride out the pre-hearing period for a very long time. These extreme values may also represent data errors. We're not sure.

Figure 14

Active ECB violations are violations that still need to be addressed, whereas resolved ECB violations have either been cured, paid, or dismissed. The above graph shows the breakdown of ECB violations, by status, which have been heard as of 2019 (we are grouping by hearing date).The dash line represents total amount of active ECB violations to total ECB Violations (active-to-total).

Active-to-total has held at at least 20% since 2001, and hit 30% in 2018, highlighting that the city has been kept busy with all the recently issued violations.

Figure 15

We were interested in assessing whether, as it did when broken out by violation status, paid-to-imposed showed some variability across penalty size (in dollars). So, we calculated the paid-to-imposed ratio across 6 different imposed-penalty bins. These were (in dollars):

  • 0-500
  • 500-1,000
  • 1,000-5,000
  • 5,000-10,000
  • 10,000-50,000
  • 50,000-100,000

and we looked at this data over the past 4 years. From 2016 to 2019, paid-to-imposed ratios decreased as the imposed-penalty bin boundaries went higher. Paid-to-imposed ratios - even in 2018 - were close to 80% for the two lowest bins. However, these ratios plummeted in the higher bins, falling to 20% and below - even in 2019 - for the two highest bins.

It would appear that the city has trouble collecting on its steeper penalties.

Figure 16

ECB Violation Type Year Normalized Violation Counts Normalized Imposed Penalty Paid to Imposed
Boilers 2017 0.04 0.02 0.43
2018 0.03 0.01 0.43
2019 0.02 0.01 0.34
Construction 2017 0.74 0.78 0.45
2018 0.77 0.80 0.34
2019 0.77 0.76 0.22
Cranes and Derricks 2017 0.00 0.01 0.83
2018 0.01 0.01 0.81
2019 0.00 0.01 0.41
Elevators 2017 0.09 0.04 0.72
2018 0.07 0.03 0.49
2019 0.06 0.03 0.43
Local Law 2017 0.00 0.01 0.66
2018 0.04 0.01 0.49
2019 0.00 0.01 0.51
Plumbing 2017 0.03 0.02 0.58
2018 0.01 0.01 0.46
2019 0.01 0.01 0.30
Public Assembly 2017 0.00 0.00 0.80
2018 0.00 0.00 1.04
2019 0.00 0.00 0.06
Quality of Life 2017 0.01 0.03 0.43
2018 0.01 0.04 0.30
2019 0.02 0.09 0.07
Signs 2017 0.01 0.01 0.39
2018 0.01 0.02 0.36
2019 0.01 0.01 0.27
Site Safety 2017 0.00 0.00 1.12
2018 0.00 0.00 0.47
2019 0.01 0.01 0.20
Unknown 2017 0.07 0.08 0.60
2018 0.04 0.09 0.60
2019 0.09 0.12 0.39
Zoning 2017 0.01 0.00 0.66
2018 0.02 0.01 0.67
2019 0.01 0.00 0.57

This table shows the percentages of DoB ECB violations:

  • issued (in counts, as grouped by issuance year)
  • imposed (in dollars, as grouped by hearing year)

as well as the paid-to-imposed ratios, also as grouped by hearing year. All data displayed is broken out by year and by violation type, for 2017 - 2019.

We can see that Construction violations account for the lion's share of both the number of issued penalties and the total dollar-value of imposed penalties. Additionally, we can see that Construction violations show a poor paid-to-imposed ratio. This relative lack of payment, paired with outsized contributions to Normalized Imposed Penalty, has led to a large percentage of the violation dollars imposed in recent years not being paid.

More quantitatively, with a paid-to-imposed ratio of just 34% in 2018 but an 80% contribution to all penalty dollars imposed, Construction penalties in 2018 single-handedly account for the current (as of our analysis) non-payment of 73% of all violation dollars imposed that year.

Quality of Life violations, with regard to paid-to-imposed ratio, tumbled even more drastically - only 7% of those that were issued last year have been paid. These account for only a small percent of the total dollar-value of imposed penalties, although that contribution grew noticeably between 2018 and 2019.

The large increase in the issuance of Construction violations (as show in figure 7), as construction activity overall has increased, supports the idea that some landlords and developers treat violation receipt as simply a cost of doing business. When breaking the data out by violation type, we see (in certain categories) declines in penalty payment severe enough to suggest that some landlords and developers may even look at these violations as something along the lines of a pay-as-you-wish cost of doing business.

It's certainly possible that some of the particularly egregious, still-outstanding balances will be remedied in the near future. However, the observable declines in penalty payment could indicate a real and negative change in the city's collection capabilities. Such a change could be driven by some combination of a number of factors, including (for example):

  • an over-burdened ECB
  • overly cheap penalties
  • developer and / or landlord malfeasance
  • developer and / or landlord financial strife

It is our hope that this analysis will be taken further (whether by some other party, by NewTrails, or both) so that the residents of New York City can continue to gain insight into the underlying causes of the deteriorating quality of life that so many of us have experienced in recent years.

Figure 17

We grouped (non-ECB) DoB violations by issued year and month from 2000 to 2019 to allow us to understand the trend in the issuance of these violations over the years.

As has already been mentioned, the periodic spikes probably coincide with auto-generated violations that are issued from time to time. That doesn't mean that this chart didn't look surprising to us! Something like yearly autocorrelation (as we saw above) was to be expected - but clusters around the same calendar month each year, that was surprising.

We were unable to come to a conclusion on whether a default on a DoB-issued ECB violation yields a DoB violation - this is an important question that we're hoping someone can answer for us. If that is the case, the jagged temporality of this data could be aggravating efforts to enforce payment on DoB-issued ECB violations, as perpetrators might be either confused about the consequences for penalty default, or aware of the fact that further penalties may be more tied to the calendar month than the date of default.

This last clause, that "further penalties may be more tied to the calendar month than the date of default," would (if true) indicate some room for improvement in this default-penalization process.

Figure 18

We wanted to put some of the normalized values we've been discussing into a dollar-value context, and so we plotted, in dollars and by hearing date year:

The DoB's expense data was only available for 2002 - 2017 in the dataset that we chose to use, and so we restricted our plot to this time range.

It is surprising to see that penalties imposed exceeded the total expense of the DoB from 2009 to 2015. For 2010 and 2011, penalties unpaid also exceeded the DoB's annual expenses.

The DoB's expenses have increased since 2014, during which time penalties imposed (in dollars) have oscillated, and were up only a small bit as of 2017, our most recent year. Unpaid penalties for 2017 are down from where they were a couple of years prior, but still tie out, proportionately, to 42% of the DoB's total annual expenses.

The DoB could potentially get much more funding for itself, and therefore many more resources with which it could squash problematic activity, if it was better able to collect on outstanding payments.

^

Recommendations

Successfully executing on penalty collection is critical to the New York City government's protection of its residents' quality of life. As we have already discussed, a failure to execute can lead to more corner cutting, unsafe conditions - more in-violation activity.

Rigorous enforcement of DoB-issued ECB penalty collection would likely reduce the number of violations in default, thus:

  • reducing the outstanding balance owed to the city
  • increasing the likelihood of penalty payment
  • discouraging bad behavior that may be driven by the opportunity for low-cost penalty default

Additionally, the maintenance of a reasonable gap between issuance and hearing dates for DoB-issued ECB penalties would reduce the ability of harmful actors to delay, e.g. until units in a new building are sold, penalty payment in such a way that is helpful to their cash flows.

The achievement of these goals may require additional human resources, since it seems that the increases in penalty-collection laxness are strongly related to increases in penalty-issuance volumes. That said, automated and possibly statistical tools could allow for a more efficient and manageable path to execution.

Finally, the city needs to ensure that violation penalties do not remain stagnant. New York City, even in its ups and downs, remains a competitive real state market. If the profits arising from real estate construction and management outpace the costs associated with violating the building code, then it is simply good business to violate the code. The penalties for violating the building code should be steep enough to dissuade real estate participants from engaging in such violations.

^

Next Steps

We can reiterate, once again, that we have painted in broad strokes. Construction activity varies from neighborhood to neighborhood, and further analysis at the borough and neighborhood levels would be very interesting to see. It would also be helpful to see measures of population density compared to construction activity, as the various "effects" of construction activity are surely felt differently, depending on how dense the surrounding area is.

Additionally, analysis on the relationships between economic factors (e.g. REIT valuations, residential rental prices, residential sales prices) and the data we've analyzed here would be interesting to see. For instance, a capEx-heavy industry such as real estate will surely exhibit cyclical tendencies - we did in fact see this in the DoB data*. Comparing these tendencies to pricing trends, an understanding of which could benefit developers's sense of market timing, was something that we'd hoped to do, but didn't have time for. And, regarding ECB violations - it would be interesting to flesh out whether there is, or whether there should be, any relationship between the real estate market's economic health and the city's penalty imposition or collection.

Finally, our initial and ambitious hope for this project was to understand if there were certain, key developers or landlords who were regularly flouting the building code and who had thus far managed to escape real punishment and / or exclusion from the NYC real estate landscape. A lack of punishment for persistent rule flouting essentially green-lights violating the building code in the name of cost cutting or project expediency. Taken further, a lack of enforcement effectively implements rule flouting as a de facto requirement for any developer or landlord who wishes to seriously compete in the marketplace.

As such, we had envisioned trying to link violations to their perpetrating landlords or developers, and to the permit applications that ultimately led to the problematic behavior. Setting aside the various obfuscations practiced by many landlords and developers (including the use of LLCs and / or shell companies) and considering only the data that we had access to, the permit application and issuance datasets did not directly (i.e. at the level of a record) link to either complaints or violations. Even complaints and violations did not directly link to each other. Executing on a "record linkage" project is well within the scope of our abilities but was not something that we were able to invest the time into completing. That said, we had been of the mind that such a linkage could have enabled us to answer questions like:

  • what types of projects are most prone to in-violation activities?
  • which permit applicants can be traced to the most violations?
  • how do the answers to each of the above questions change over time?

Answers to such questions would better arm the city to weed out problematic actors from its buildings ecosystem.

^

Appendix

Figure 1 Data

  • Growth in Yearly Figures: Applications, Permits, Complaints, and Violations
  • * in below table indicates estimated value, due to the data issues explained here
  • total annual applications from 2017 to 2019 were estimated based on the applications-to-permits ratio in 2016
  • back to figure
Year Applications Permits Complaints Violations (ECB)
2000 67,735 101,496 43,520 41,030
2001 67,320 105,728 45,124 42,059
2002 70,045 111,931 51,991 40,566
2003 72,817 122,595 62,542 45,041
2004 80,833 137,618 78,212 44,962
2005 86,264 150,087 108,873 49,196
2006 86,896 155,861 112,953 51,687
2007 90,201 158,250 129,247 58,225
2008 84,548 150,370 142,991 76,940
2009 70,726 135,770 127,278 78,540
2010 74,889 136,243 113,516 69,054
2011 77,847 143,943 103,800 57,781
2012 81,843 148,534 100,377 47,173
2013 93,195 161,322 95,242 45,773
2014 107,943 170,787 101,680 52,095
2015 107,544 182,714 101,984 52,582
2016 109,527 191,006 114,679 59,838
2017 113,287* 197,556 126,587 79,105
2018 112,290* 195,817 129,562 88,482
2019 110,330* 192,399 142,873 92,800

--

Figure 2 Data

Lags Permits Complaints Violations (ECB)
1 -0.382373 -0.328890 -0.291068
2 -0.008993 0.192306 0.025308
3 0.146137 0.039626 0.068256
4 -0.259690 -0.244665 -0.107328
5 0.148509 0.080146 0.099334
6 -0.182140 -0.225224 -0.139810
7 0.089662 -0.070911 0.056295
8 -0.176381 -0.075883 -0.116743
9 0.147122 -0.034335 0.122574
10 -0.159407 -0.006957 -0.069506
11 -0.143229 -0.026291 -0.085073
12 0.707353 0.449963 0.236153
13 -0.362506 -0.235264 -0.130263
14 0.162864 0.336698 0.134334
15 0.026297 -0.056239 -0.007688
16 -0.247162 -0.130894 -0.096528
17 0.242613 0.078137 0.010223
18 -0.336469 -0.339339 -0.064154
19 0.134917 -0.011767 0.108028
20 -0.121456 -0.054585 -0.180632
21 0.049985 -0.116279 0.093528
22 -0.115574 0.184815 -0.054410
23 -0.040642 -0.113334 -0.011686
24 0.499464 0.403487 0.236234
25 -0.177592 -0.105215 -0.137769
26 0.118807 0.337581 0.008371
27 -0.060310 -0.130789 -0.012436
28 -0.106392 -0.116708 0.004337
29 0.187828 0.098104 0.079013
30 -0.370341 -0.357268 -0.146401

--

Figure 3 Data

Year applications permits
2000 0.00000 0.000000
2001 -0.61268 4.169622
2002 3.410349 10.28119
2003 7.502768 20.78801
2004 19.337123 35.58958
2005 27.355134 47.87479
2006 28.288182 53.56368
2007 33.167491 55.91747
2008 24.821732 48.15362
2009 4.415738 33.76881
2010 10.561748 34.23484
2011 14.928767 41.82135
2012 20.828228 46.34468
2013 37.587658 58.94419
2014 59.360744 68.26968
2015 58.771684 80.02088
2016 61.699269 88.19066
2017 67.251284 94.64412
2018 65.779044 92.93075
2019 62.885358 89.56313

--

Figure 4 Data

Year Construction Equipment Equipment Work Plumbing
2000 13.569 44.797 20.807
2001 14.418 44.733 21.625
2002 16.969 44.827 22.798
2003 19.812 45.836 25.495
2004 23.701 50.635 27.585
2005 28.524 53.444 29.536
2006 32.166 53.891 30.14
2007 34.315 58.221 29.496
2008 29.978 60.043 30.445
2009 24.86 57.115 29.677
2010 24.361 59.017 28.937
2011 25.051 66.12 30.004
2012 26.38 68.324 29.076
2013 28.199 75.961 31.998
2014 28.746 80.712 33.718
2015 29.4 85.193 35.743
2016 30.295 91.03 36.585
2017 29.511 92.948 38.22

--

Figure 5 Data

  • Yearly Ratios, Over Time: Complaints, Violations; to Permits
  • back to figure
Year Complaints to Permits Violations (ECB) to Permits
2000 0.428785 0.404252
2001 0.426793 0.397804
2002 0.464492 0.362420
2003 0.510151 0.367397
2004 0.568327 0.326716
2005 0.725399 0.327783
2006 0.724703 0.331622
2007 0.816727 0.367930
2008 0.950928 0.511671
2009 0.937453 0.578478
2010 0.833188 0.506844
2011 0.721119 0.401416
2012 0.675785 0.317591
2013 0.590384 0.283737
2014 0.595361 0.305029
2015 0.558162 0.287783
2016 0.600395 0.313278
2017 0.640765 0.400418
2018 0.661648 0.451861
2019 0.742623 0.482331

--

Figure 6 Data

Year Illegal Conversion Permit – None Elevator (Old Category) Elevator (New Category)
2006 19.424 15.571 9.281 NaN
2007 22.739 18.483 9.413 NaN
2008 21.966 19.855 9.431 NaN
2009 23.409 16.396 8.250 NaN
2010 18.370 18.454 7.082 NaN
2011 15.745 19.580 6.433 NaN
2012 13.977 15.190 7.002 NaN
2013 14.273 12.143 7.520 NaN
2014 14.568 13.508 8.034 NaN
2015 13.527 13.366 8.952 NaN
2016 14.622 15.110 11.572 NaN
2017 17.179 15.566 11.992 NaN
2018 17.107 14.370 NaN 9.674
2019 15.603 15.916 NaN 10.643

--

Figure 7 Data

Year Construction Elevators Quality of Life Unknown
2006 25.230 6.217 5.475 2.040
2007 26.208 7.923 5.385 4.113
2008 40.431 10.373 3.161 9.052
2009 45.217 7.888 0.627 15.514
2010 40.791 5.757 0.604 14.334
2011 31.625 4.972 0.363 15.344
2012 24.287 4.065 0.163 14.305
2013 24.700 4.276 0.252 13.566
2014 30.764 4.216 0.429 12.234
2015 38.801 3.957 0.633 4.648
2016 44.424 6.653 0.807 2.095
2017 58.225 6.925 1.058 5.405
2018 67.900 6.088 0.922 3.980
2019 71.532 5.439 1.491 8.614

--

Figure 8 Data

Issue Year Mean Median Hearing Year Mean Median
2000 822.829466 250.0 2000 843.5884788478850 250.0
2001 692.964526 200.0 2001 678.3566753627170 180.0
2002 729.038357 250.0 2002 745.3990105936440 250.0
2003 700.333407 250.0 2003 701.9465517621390 240.0
2004 802.410591 350.0 2004 762.2334090558210 250.0
2005 957.911172 350.0 2005 881.2444449247800 350.0
2006 1138.404550 480.0 2006 1058.6154163742200 400.0
2007 1175.498205 500.0 2007 1149.1942861513700 500.0
2008 1308.193930 500.0 2008 1210.7764213892600 500.0
2009 1859.663222 500.0 2009 1542.9416647181100 500.0
2010 2380.383830 500.0 2010 2315.5780727143900 800.0
2011 2620.992368 800.0 2011 2546.2986676269400 800.0
2012 2661.495654 500.0 2012 2497.6836732759600 500.0
2013 2310.251285 500.0 2013 2416.036066691030 500.0
2014 2367.043118 800.0 2014 2349.4754807673100 800.0
2015 2366.361085 800.0 2015 2459.605247914740 800.0
2016 1871.004867 800.0 2016 1956.63924582552 800.0
2017 1762.884373 800.0 2017 1710.841972509570 600.0
2018 2387.357734 800.0 2018 2134.2540692700700 800.0
2019 2622.369828 1250.0 2019 2476.1035762572300 800.0

Figure 9 Data

Issue Year Mean Median Hearing Year Mean Median
2000 420.31777187424 0.0 2000 418.9148069949860 0.0
2001 392.00367364892200 0.0 2001 379.6388663096720 0.0
2002 418.0449115022440 0.0 2002 430.89797971217300 0.0
2003 404.7831588996700 0.0 2003 411.7189105375680 0.0
2004 449.83422356656800 130.0 2004 426.1413630047870 130.0
2005 524.0713415724850 130.0 2005 488.05953224969800 130.0
2006 553.7166212006900 180.0 2006 537.31712831840200 130.0
2007 570.222657277802 180.0 2007 551.35253317780200 180.0
2008 653.3622019755650 200.05 2008 568.10924556440300 180.0
2009 817.5204583651650 200.0 2009 696.82715263704900 200.0
2010 865.5988876821020 100.0 2010 896.39615858518100 250.0
2011 847.4791803534040 0.0 2011 897.44453510983300 200.0
2012 835.1980980645720 0.0 2012 835.44033324487500 0.0
2013 822.7347117296220 200.0 2013 837.3819082837040 100.0
2014 893.4595888281020 250.0 2014 882.41543356069800 250.0
2015 947.3545804647990 400.0 2015 966.5381605267200 400.0
2016 894.4359687823770 250.0 2016 918.3032983302060 250.0
2017 819.444093040896 0.0 2017 818.7907139010900 0.0
2018 748.740919622064 0.0 2018 795.53807542021900 0.0
2019 305.1469738146550 0.00 2019 579.71646015236200 0.0

Figure 10 Data

  • ECB Penalties Yearly: Number Issued, by Issue Date; Paid-to-Imposed Ratio, by Hearing Date
  • back to figure
Year Ratio (%) Violations (K)
2000 49.658668 41.030
2001 55.964491 42.059
2002 57.807694 40.566
2003 58.653883 45.041
2004 55.906938 44.962
2005 55.382991 49.196
2006 50.756593 51.687
2007 47.977312 58.225
2008 46.921069 76.940
2009 45.162248 78.540
2010 38.711549 69.054
2011 35.245061 57.781
2012 33.448604 47.173
2013 34.659329 45.773
2014 37.557975 52.095
2015 39.296474 52.582
2016 46.932683 59.838
2017 47.858933 79.105
2018 37.274759 88.482
2019 23.412447 92.800

Figure 11 Data

  • Normalized Penalty Imposed Vs. Paid-to-Imposed Ratio, by Hearing Date, 2010 - 2020

  • back to figure

  • Normalized Penalty Imposed

Year DEFAULT IN VIOLATION PENDING
2010 0.2551 0.2801 0.0000
2011 0.2781 0.3061 0.0000
2012 0.6381 0.3448 NaN
2013 0.6153 0.3599 0.0000
2014 0.5849 0.3880 NaN
2015 0.5691 0.4068 NaN
2016 0.5547 0.4058 NaN
2017 0.5055 0.4449 0.0015
2018 0.5220 0.4393 0.0000
2019 0.5598 0.4103 0.0002
  • Paid to Imposed Ratio
Year DEFAULT IN VIOLATION PENDING
2010 0.473319 0.838798 0.000000
2011 0.322927 0.765718 NaN
2012 0.129904 0.661808 NaN
2013 0.132598 0.651158 NaN
2014 0.151629 0.654419 NaN
2015 0.173058 0.652355 NaN
2016 0.203692 0.764692 NaN
2017 0.174569 0.753715 0.663402
2018 0.072933 0.668929 NaN
2019 0.019203 0.480588 0.022704

Figure 12 Data

  • Average Days for ECB to Hear a Violation: 2000 - 2019
  • back to figure
Year Mean Median
2000 140.609901 74
2001 159.895876 90
2002 162.271387 70
2003 150.907231 70
2004 163.295384 68
2005 149.485777 63
2006 144.881281 63
2007 161.831988 67
2008 159.534562 62
2009 184.770590 74
2010 195.284887 87
2011 185.827627 83
2012 147.744000 61
2013 141.319524 56
2014 121.419419 58
2015 120.036203 56
2016 125.854862 60
2017 121.893496 68
2018 137.364766 86
2019 145.701082 105

Figure 14 Data

  • Active and Resolved ECB Violations as of 2019, by Hearing Date
  • back to figure
Year Active Resolved Active to Total (%)
2000 3.228 22.966 12.32343284721690
2001 4.864 36.838 11.663709174619900
2002 4.985 34.199 12.722029399755000
2003 5.275 39.635 11.745713649521300
2004 5.585 39.426 12.408078025371600
2005 6.124 39.98 13.283012319972200
2006 7.175 41.188 14.835721522651600
2007 8.572 43.698 16.39946431987760
2008 11.047 53.985 16.98702177389590
2009 12.52 64.457 16.264598516440000
2010 14.146 64.896 17.896814351863600
2011 14.177 55.241 20.42265694776570
2012 10.77 41.983 20.415900517506100
2013 9.454 37.056 20.326811438400300
2014 10.361 39.63 20.725730631513700
2015 12.11 39.679 23.38334395334920
2016 13.87 42.122 24.77139591370200
2017 18.659 51.903 26.443411467928900
2018 26.092 60.397 30.167998242551100
2019 32.187 54.054 37.32215535534140

Figure 15 Data

Hearing Year (0, 500] (500, 1000] (1000, 5000] (5000, 10000] (10000, 50000] (50000, 100000]
2016 89.959553 87.277110 63.722627 41.092119 19.575326 16.12171
2017 86.242135 81.987317 59.067698 43.057869 21.543053 3.979885
2018 81.402043 77.379626 61.629957 27.893900 12.103294 4.879977
2019 67.176280 65.304715 48.510967 17.165930 4.793826 0.449684

Figure 17 Data

Date (2000 - 2009) Issuances (k) Date (2010 - 2019) Issuances (k)
2000-1 2.123 2010-1 2.690
2000-2 33.905 2010-2 2.576
2000-3 5.972 2010-3 4.211
2000-4 2.193 2010-4 4.916
2000-5 2.546 2010-5 4.319
2000-6 2.603 2010-6 4.946
2000-7 2.243 2010-7 5.077
2000-8 2.683 2010-8 4.835
2000-9 2.225 2010-9 4.981
2000-10 2.532 2010-10 10.466
2000-11 1.861 2010-11 5.184
2000-12 1.685 2010-12 7.405
2001-1 2.449 2011-1 6.142
2001-2 34.193 2011-2 4.131
2001-3 6.380 2011-3 4.874
2001-4 3.300 2011-4 4.309
2001-5 3.172 2011-5 4.740
2001-6 3.178 2011-6 5.733
2001-7 2.752 2011-7 5.183
2001-8 2.854 2011-8 5.535
2001-9 1.736 2011-9 4.844
2001-10 2.314 2011-10 4.756
2001-11 1.995 2011-11 4.885
2001-12 1.950 2011-12 33.762
2002-1 2.549 2012-1 7.807
2002-2 5.907 2012-2 4.969
2002-3 34.725 2012-3 4.796
2002-4 1.782 2012-4 4.183
2002-5 1.966 2012-5 7.824
2002-6 1.575 2012-6 3.891
2002-7 1.729 2012-7 3.756
2002-8 6.001 2012-8 6.442
2002-9 1.450 2012-9 3.931
2002-10 2.181 2012-10 9.046
2002-11 1.797 2012-11 3.596
2002-12 1.939 2012-12 3.546
2003-1 2.357 2013-1 8.039
2003-2 1.519 2013-2 3.929
2003-3 35.990 2013-3 22.026
2003-4 1.904 2013-4 3.803
2003-5 1.970 2013-5 3.862
2003-6 2.252 2013-6 5.713
2003-7 2.181 2013-7 4.694
2003-8 2.297 2013-8 6.075
2003-9 2.569 2013-9 4.409
2003-10 2.561 2013-10 4.618
2003-11 2.015 2013-11 3.551
2003-12 2.323 2013-12 4.939
2004-1 2.809 2014-1 5.804
2004-2 31.794 2014-2 3.417
2004-3 2.881 2014-3 13.461
2004-4 6.278 2014-4 3.982
2004-5 2.078 2014-5 39.463
2004-6 2.392 2014-6 5.096
2004-7 1.830 2014-7 4.481
2004-8 2.185 2014-8 3.479
2004-9 2.569 2014-9 3.829
2004-10 2.721 2014-10 10.635
2004-11 6.155 2014-11 3.662
2004-12 2.698 2014-12 4.469
2005-1 30.441 2015-1 2.780
2005-2 2.739 2015-2 3.261
2005-3 3.263 2015-3 22.767
2005-4 3.797 2015-4 3.530
2005-5 3.671 2015-5 4.936
2005-6 3.533 2015-6 10.186
2005-7 2.922 2015-7 4.228
2005-8 3.662 2015-8 4.492
2005-9 3.188 2015-9 3.300
2005-10 7.880 2015-10 18.763
2005-11 2.811 2015-11 4.159
2005-12 2.685 2015-12 4.091
2006-1 29.530 2016-1 4.141
2006-2 2.848 2016-2 5.454
2006-3 3.111 2016-3 5.441
2006-4 2.435 2016-4 4.433
2006-5 3.072 2016-5 4.399
2006-6 3.470 2016-6 11.082
2006-7 3.616 2016-7 3.842
2006-8 4.221 2016-8 5.005
2006-9 4.380 2016-9 3.931
2006-10 4.596 2016-10 3.388
2006-11 4.786 2016-11 18.943
2006-12 3.649 2016-12 10.193
2007-1 31.684 2017-1 18.253
2007-2 3.541 2017-2 22.102
2007-3 4.939 2017-3 3.955
2007-4 8.444 2017-4 3.274
2007-5 4.312 2017-5 5.571
2007-6 5.269 2017-6 4.494
2007-7 4.442 2017-7 4.046
2007-8 5.107 2017-8 4.788
2007-9 5.060 2017-9 10.928
2007-10 5.140 2017-10 4.157
2007-11 10.118 2017-11 4.720
2007-12 4.617 2017-12 4.084
2008-1 32.876 2018-1 17.802
2008-2 5.238 2018-2 4.370
2008-3 11.259 2018-3 5.077
2008-4 6.257 2018-4 23.655
2008-5 6.160 2018-5 6.861
2008-6 5.919 2018-6 4.272
2008-7 5.457 2018-7 5.544
2008-8 4.910 2018-8 3.754
2008-9 4.543 2018-9 3.565
2008-10 4.857 2018-10 4.149
2008-11 3.940 2018-11 3.090
2008-12 3.982 2018-12 2.711
2009-1 31.115 2019-1 4.668
2009-2 6.654 2019-2 4.457
2009-3 3.607 2019-3 6.024
2009-4 3.440 2019-4 11.889
2009-5 3.129 2019-5 13.596
2009-6 3.662 2019-6 4.210
2009-7 3.509 2019-7 4.242
2009-8 3.250 2019-8 7.003
2009-9 3.159 2019-9 16.927
2009-10 2.996 2019-10 6.355
2009-11 2.525 2019-11 21.825
2009-12 29.633 2019-12 3.924

Figure 18 Data

  • Penalties Imposed, Penalties Unpaid, and DoB Expenses: Yearly, by Hearing Date
  • All numbers are in Millions
  • back to figure
Year Unpaid Penalties Imposed Penalties Total Expense
2002 12.587589 29.833850 56.378
2003 13.167918 31.848017 57.862
2004 15.238749 34.560425 57.895
2005 18.190307 40.769893 64.273
2006 25.252732 51.281448 81.526
2007 31.263537 60.095966 87.195
2008 41.800358 78.751320 99.410
2009 65.133895 118.77564 109.67
2010 112.18491 183.04413 101.53
2011 114.47169 176.77678 98.974
2012 87.689985 131.76280 95.062
2013 73.427941 112.37708 102.48
2014 73.342733 117.45732 98.679
2015 77.328931 127.38787 108.67
2016 58.141621 109.56201 134.72
2017 62.952057 120.73411 150.35