Tip:
Highlight text to annotate it
X
So as Andy said, my name is Dave Kamper and I'm one of your business agents here at MAPE
and before I get started on the data itself, I want to talk a little bit about why this
kind of project came to be. I belong to what you might call the crisis school of the labor
movement, where I believe that if we don't change and soon, we face extinction within
a couple of generations. And the labor movement has to do whatever it takes to be successful
because we're the best voice for working families out there. It was about 2004, right at the
beginning of this analysis, that I came across what I seriously consider the best book ever
written about how to solve the problem of the American labor movement. And I promise
you this is not a joke. The best book ever written on how to solve the problems of the
American labor movement is Money Ball by Michael Lews. I would encourage all of you to take
the time to go read it. You could see the movie too, but the book is what you need to
read. Let me explain why it's important. The book is the story, for those of you who aren't
familiar, of Billy Beane the general manager of the Oakland Athletics and how the Oakland
Athletics, who have the lowest payroll of any team in Major League Baseball, were able
to compete successfully against the New York Yankees who had the greatest payroll of any
team in MLB. That is to say Moneyball is a study about what you do when your opponent
has vastly more resources than you do. That ought to sound really familiar because what
the book is about is strategic resource allocation. It's about figuring out what to do when you
have a situation where what you have isn't even sufficient to handle what you want to
do, let alone what you need to do. What do you do when that happens? And that's one of
the things that as I was reading Moneyball 10 years ago that caught my interest about
it. Another thing that I think makes the book so valuable in our study of the labor movement
is that there's so much in common with our situations. Both baseball and the labor movement
started about the same time in American history. Both reached their peaks in the 1930s with
Babe Ruth and the Wagner Act. Both have lots of troubles in the 80s and 90s. Both face
uncertain futures today. In addition to the kind of work we're talking about - the work
I'm interested in here and the work of how unions do what unions do - because what unions
do requires professional skills, just like baseball and requires experience, just like
baseball. Just like Billy Beane was struggling with, we lack a formal certification system
to train our people. If you want to be a lawyer, you go to law school. You want to be a doctor,
you go to medical school. You get certified, you get past your boards. Even if you want
to be a land surveyor, you've got to get your certification and go to school. Even cosmetology
has a school for it, for people who want to be certified in what they do. And of course,
those of us who do this know the people who sit on the other side of the table from us,
have a certification - a professional in human resource certification. Which these days is
de rigueur for anyone looking for a senior human resources job anywhere in management?
In the labor movement, we have no equivalence. We have some labor education programs, and
many of them are very excellent, but the truth is, the vast majority of people who are running,
leading and part of the labor movement today have never had any kind of formal training
beyond what they got from the previous generation. How to we train our stewards? We bring in
stewards who have done it for a while and train them. How do we train negotiators? We
bring in previous negotiators. New officers are trained by old officers. New staff is
brought up to speed by old staff. This is how we do things in the labor movement. And
how they did things in baseball. Where conventional wisdom and untested assumptions are what drives
things. For decades in baseball, the presumption was that in certain situations you had to
bunt. Turns out that's not true but that's what everyone got taught. And we too in the
labor movement have untested assumptions that drive us and part of having those untested
assumptions is that too often success becomes following the conventional wisdom. Well, we
did what we were supposed to do - we put in the left-handed pitcher and so even though
we didn't win the game, we did what we were supposed to do. We in the labor movement do
that as well. We follow the steps we're supposed to follow and its results didn't come out
the way they did, well ok. We've got to change the way we go about thinking about it - success
should be success. Success should be doing good things that help improve the lives of
our members, the lives of America's working people in the rest of the world. That's what
success should be. In order to do that, again, I want you to go read Moneyball about this.
What you do is you assume nothing, you don't rely on conventional wisdom, you use objective
measures whenever possible, and remember possible is an important word there, you test theories
and you experiment. If you read Moneyball, that's what that book
is about and that's what we're also talking about today. Because MAPE spent the money
on a new database system called Unionware, we have the ability to sort through data much
better than we ever have before. So this presentation would have taken 10 times as long to prepare
under our old system. As a result, I was able to do this in much less time. What we're going
to look at today is data. Data from our database on grievances in MAPE. Whatever trends and
problems there might be; and hopefully in our discussion we can talk about recommendations.
I want to begin with some caveats. First, this is really important. Not everything that
can be measured is important and not everything that's important can be measured. I can't
stress this enough. Just because you have something in the numerical form or in a piece
of data you can measure, doesn't mean it's always more important than the thing you can't
quite measure. And just because you can't measure it doesn't mean it can't be really
vital. And so while we're going to talk about hard numbers today, and statistics, that's
not the whole story. It's only part of it. Another piece that you need to remember is
that data is not the plural form of anecdote. You're going to see data presented to you
today. And all of you are going to say 'Oh right, that seems like a situation I had in
such and such a place ..." That's a natural human tendency to go there. It's what we do;
we float to our own experience through what we see. But we have to be careful as a union
not to let those individual anecdotes drive us when the data might say something different.
On the flip side of that, of course correlation is not causation. Just because we see a link
between things doesn't mean one caused the other. It requires much deeper exploration
to really understand what might be happening in certain situations. And then as a historian,
there's no way I can't bring up my old pal George Santayana.
If we learn things about our past, and we don't do anything about it, then it's our
own fault what happens to us is my point. Let me talk about the methodology here. Pulling
up from 2004 to 2013 and this is - I went to the middle of 2013 to June 30th - we've
approximate 1,500 grievances. And we can sort that data by date, by agency, by local, by
type of grievance. Grievances of course are a tricky thing. Sometimes a grievance isn't
a bad thing. Sometimes a grievance is just a step in the process. But overall I think
it is safe to say that a grievance is representative of a few different things - one - the state
of a labor management relationship; two - the union's activism and tracking down and finding
grievances; and three - management's reliance of or ignorance of the collective bargaining
agreement. And so in general, I think when you have more grievances, that's a sign of
trouble. When you have less grievances, that's a sign that things might be going better.
There are exceptions to that rule aplenty. But I think in general that shows. Now the
key takeaways from this, first of all, grievances are up. I'll show you how much. The increase
is tied to very specific areas of MAPE and we'll talk about that. And there's something
very interesting going around with grievances and union activists, which I will get to near
the end of my presentation and I'll talk about. So first of all, over time, grievances are
going up. The red line shows our grievances per member per year. In 2004, we had about
10,500 members. In 2013 we had about 13,000 - so we would expect grievances to rise, just
by having more members, so this is weighted for membership growth. This is grievances
per member. Now you'll notice there's another line here. And let me take a moment to explain
the statistical measures I'm using here. I'm not going to try to teach anyone statistics
- I don't know how. But I want to explain for a moment what we're doing here. This green
line is what's called a regression line. It's basically a statistically calculated way of
saying "Is this bumpy red line actually going in a direction or is it just a bumpy line?"
Because the human eye can be deceived by these things; it can look to us like there's a trend
but there might not be. The R-squared number is the statistical measure of how important
that line is. Now my political scientist brother-in-law says .3594 is pretty significant and means
that grievances are in fact going up in time not that is just looks like it.
What are we grieving? See slides. Those are the articles of the contract that
we grieve the most. That's not to say that's where most of the violations occur, but that's
where we grieve the most. The fact that grievances are up over time doesn't necessarily mean
that violations are up over time either. But I also want to talk about where they happen.
This is an entirely useless set of numbers. This is how many grievances in various state
agencies from 2004 to the summer of 2013. As you know, this is an entirely useless number
because some of these agencies are 200 people and some of these agencies are 1,500 people
or more. And so looking like this may look alarming at certain places and not alarming
at others. The truth is it's almost entirely useless. Thankfully again we have a statistical
tool that can help us figure this out. What I figured out was something called the Z score
which basically looks at if we assume that the grievances were averagely (that's not
a word, but we're going to use it) spread out amongst all the agencies, how many would
each agency have? And comparing that to where they really are - how far above or below the
predicted average they are and we get the number that's called a Z score, which is almost
always between 2 and minus 2 and if it's positive, there's more than you should expect and if
it's negative, it means there's less than you should expect. If you break it down by
Z scores, this is what it looks like. The Dept. of Labor and Industry shoots way up
over 2.0 - Human services is next - I think people might have predicted that. But departments
where we have a fair number of grievances like public service, pollution control agency-
they're below average and several of them are pretty close to the norm like the Minnesota
State Colleges and Universities system, Dept. of Transportation, Education - they're pretty
close to the norm. This however is also at times a misleading graph. And I want to look
at the department of labor and industry - from 2004 to 2008 they had 55 grievances in DOLI.
Since then? 8 grievances. So overall, from 2004-2013 their Z score is through the roof,
but if you break it down over time, something great has happened over there I think. I don't
know that. It might just be chance, but it's not. Something probably has happened to change
the labor-management dynamic in DOLI. And it started around 2009. And it's made things
better there so the number of grievances drops off to almost nothing. So let's look at where
the trouble spots are - (see slides) That's about
what you'd expect based on the size of those agencies. But if you go to 2011-2013, all
of a sudden those four agencies have taken up a huge chunk of the grievances. And in
fact, if you look at the overall trend, our grievances are up. If you take those four
agencies out, our grievances are actually down over time. All the increase is coming
from these four agencies plus another one I'll talk about briefly later that's too small
to accurately measure. We've got some serious challenges to look at here. And in fact, MnSCU's
only on here because it basically hasn't gone down. It's actually kind of leveled over time.
I'm going to look at the other three agencies where I think there's really interesting things
going on. First, the department of health. This is the number of grievances per year
once again weighted for membership growth. Look if you will at this area down here from
2007-2010 there were five or fewer grievances every year in the department of health. All
of a sudden in 2011, it shoots through the roof. The overall trend line, the R-squared
number is not terribly significant because it was higher as well in the past. So we have
two questions we have to dig into in health - one is what was going right in 2007, 8,
9 and 10, and what's gone haywire since then? And frankly, we need to answer both those
questions if we really want to understand what's going on because we want to know why
there are fewer grievances then and now why there's more.
Next I want to talk about DHS. This is their grievances per year, again weighted for membership
growth, again if you look at that trend line, it's not very big. In fact, through most of
the period, the department of human services stays fairly level in the amount of grievances
it has. It has more than the average, but stays level. However, if we break this down
to quarters of the year, starting just in 2011, this is what we see. Suddenly in middle
of 2012, it shoots up and then accelerates to astronomical levels in the beginning of
2013. This is a huge increase. It is statistically improbable to say the least that this is chance.
Something's going on over there. Here's one of those moments where we want to use anecdote
instead of data because those who are familiar with DHS know that right where things started
changing, a certain personnel figure in the HR department came on staff. That's really
interesting and again, it would be easy to try to say 'Oh it's because this person was
brought in.' but you can't know that. We just can't. There is however, even more interesting
data from DHS. This should blow your mind. This is the regions' numbers of members at
DHS and number of grievances in the last 3 years. The Lafayette building, which has 414
team members has 29 grievances. The Anderson building which has 479 MAPE represented employees
has 4 grievances. There's a test you can run called a pi squared test, which is the same
test you'd run if you threw a coin in the air and it came up heads you'd say is it weighted
or just chance? You could run that test on this and it's basically a 99.9% likelihood
that this is not random. Something's happening in the Lafayette building and something's
happening in the Anderson building because frankly that's a really low number in the
Anderson building. If you take the Anderson building out of the equation and look at the
rest of DHS, in the year 2013, the first six months of the year, there are 7.5% of the
membership and 25% of MAPE's grievances. From just DHS - not counting their biggest area.
That's really significant. Last agency I want to look at is the Department of Revenue. You
cannot in the world of statistics find a line as clear as this one, that's a steady increase.
Right up the line there, you have one year of a dip, but other than that, that's a steady
increase and an R-squared of .7633. Clearly it's going up over time. Again, it's specific
to certain areas. The Revenue central office has 70% of all our represented employees in
the revenue department. From 2011-2013, 86% of revenue department grievances were in one
area - the central office. Again, if you look outside the central office, the department
of revenue is flat or shrinking in number of grievances it has. But in the central office,
the numbers have shot through the roof. That's something we need to know about and dig into
and figure out what's going on and why. This is all kind of gloomy so let me mention a
couple of places on the good side - corrections, DEED, DOLI and DNR - have all seen huge declines
in the number of grievances they have. Good job! That's really excellent. One other place
worth noting is region 19, which is Mankato, Faribault, Albert Lea - it has also seen a
very steady growth in grievances. However, it's not clear to me how much of that isn't
the fact that the department of human services has that facility in St. peter that generates
a lot of grievances. So this actually could be the department of human services factor,
rather than region 19 factor. It's hard to disentangle those different factors.
A couple of other patterns or non-patterns of note - there's a seasonal variation to
when we file grievances. We're statistically far more likely to grieve in the summer than
we are in the winter. You can test it mathematically. It's a statistically sound difference. We
grieve more in the summer. I would have guessed the opposite myself. I would have guessed
in the summer our members were half checked out and on vacation and not filing grievances
- not true. It's not true year on year. There are certain agencies that are different, but
overall, it's a very strong trend. A couple of non- patterns to note - there's no difference
in the rate of grievances between members and fee payers. They both have about the same
rate of grievances. And there's no difference between the Twin Cities and Greater Minnesota.
I was expecting different. I was expecting that the Metro is where the bosses are, they're
watching more, they're more likely to have things going on. Not true. There's no statistically
different correlation between where you are in the state and where you're grievances are
- unless you're in the Lafayette building or Revenue Central. Now I want to get to the
most interesting statistics we have. Union activists - I mean people who are a steward,
an officer, a member of a statewide team whether it be a meet-and-confer, negotiating or a
board of directors team or anyone who attended a delegate assembly from 2011 to 2013. I only
use those years because I don't know who the activists were in 2004. From 2011-2013, that
total number is about 3.4 percent of the total MAPE membership. However, if you look at how
often they file grievances, they file grievances at about 3 times the rate as non-activists
do. In fact the exact percentage is 310%. So your first reaction is to panic. Your first
reaction is to fear that oh my god, they're clearly out to get activists; they're retaliating
and out to get us. I don't know. I want to throw some caveats out there and make us think
a little bit about it. First of all, let's look at where the grievances go. I said earlier
that 2/3s of all our grievances are in response to member discipline and about 1/3 are in
response to contract language. That is true for different regions, for different agencies,
for different times. It's not true for union activists. At 3.4% of the membership, we're
6.4% of the disciplinary grievances, but we're almost 17% of the contract grievance languages.
That is to say we're 96% more likely to file disciplinary grievances but almost 500% more
likely to file a grievance over contract language. So I want to offer a couple of hypothesis
- and they're only hypothesis and things to think about. The first one is that not all
discipline is grieved. When people receive discipline, we have a right to be there as
a union, we find out about it and encourage someone to file a grievance. They sometimes
don't. I would bet all the money in my pockets against all the money in your pockets that
activists are more likely to grieve discipline than non-activists. I suspect that's very
true. Another thing worth noting is that labor-management
friction can be a source of grievances. For example, this year we filed several grievances
over granting of union leave. Well the only people who put in for union leave are union
activists, so they're the only ones who are going to be involved in those. Also note that
some grievances are tactical or political, that we file the grievance so we can get a
conversation in a room or as part of a larger strategy for something. We may be more likely
to use an activist as a face of that grievance than otherwise.
This I think is very important - our activists are more likely to know the contract and more
likely to know what their rights are and more likely to see when something goes wrong. And
if you know the contract, you're going to be more likely to grieve the contract. On
the flip side of that, MAPE can only file a grievance when MAPE hears about it. Now,
we're not supposed to use the union as a third party, we're all the union and that's very
true. But nonetheless, grievances are filed by the business agents on our computers in
our offices downstairs. If a business agent doesn't know there's a grievance to file,
there's no grievance filed. And so again, activists know who to contact, know where
to go, know how to address a problem. I think therefore it's very likely that union activists
are more likely to file especially contract language grievances because they know where
to go and they know what their rights are. Now there's an implication there - an implication
there that I find extremely concerning. And I can't prove this to be true, because correlation
is not causation but I believe this to be true - that we're potentially missing dozens
or even hundreds of grievances every year. My belief is that the contract is being violated
far more frequently than we're aware of it. And that worries me enough to use an extraordinarily
misleading graph. If all our members grieved at the same rate as our activists, in 2012
instead of 160 grievances, we'd have over 500 grievances. Part of the things we need
to think about going forward are how are we going to be able to deal with that traffic
in grievances? If we had a doubling of the number of grievances we file, how would we
handle that? I don't know the answer to that question. What I want to avoid more than anything
in the world is the union death spiral. See slide.
That has a flip side as well. See slide. This, people, is how unions die. There's a
lot of other ways union die too, as our executive director from Wisconsin can tell us, but this
also a way unions die. We want to go the other way. A positive loop of revitalization - see
slide. The "does something about it" doesn't have to be a grievance by the way. Sometimes
a grievance isn't the best choice. The more we explore more creative options of problem
solving, the better off we are. Once members see we are advocating for them, they are more
likely to stand up for their own rights. And once management sees the union standing up,
then management is going to become more aware of the contract and less likely to violate
it. What's interesting if you think about this to the absurd level, both the death spiral
and the positive spiral end the same way - with very few grievances being filed? But in one
case, it's because our union is beaten to the ground and unable to defend itself and
in the other case, it's because we're proud and strong and they don't dare to try and
mess around with us. Quite obviously, I think we know which direction we'd like to go.
Where do we go from here? See slide. The DHS stewards have already taken some of this date
to their meet and confer and management was very surprised by it and management told them
that they had no way of calculating this data themselves so we have information the boss
doesn't and that's a good thing. But we need to explore elsewhere too. We need to go back
to DOLI and figure out what happened right 5 years ago there. And we need to go to revenue
and to health and to other areas and say "look, it looks like there's a problem here. Let's
dig into it and find out what's wrong and then decide what actions we need to take.
This is a member-run union so it's the members who will decide which actions we take. In
the face of this data, it seems to me there are things we have to do.
That's the end of my formal presentation.