Demand and Supply

One of the broadest measures of the economy is Gross Domestic Product (GDP). The Bureau of Economic Analysis (BEA) attempts to measure and calculate the parts of the economy that form the basis of the formula: GDP = C + I + G + NX. They release GDP quarterly and revise the result monthly. Then every few years, the whole calculation is revised again. There are many assumptions and estimates that take time to work through.

GDP is a measure of the consumption of the goods available to economic agents (consumers, business, and government). The largest part of the formula is C, personal consumption expenditures. This part of the formula has been about 66 to 68% of the total for a few decades. Economic policy has come to be constructed around trying to increase the size of C. The theory being that increasing the economy is most effective when policy guides an increase in C.

Another way to look at GDP is to realize that measuring consumption is equivalent to measuring demand. When you realize that concept, then the policy to affect C is based on the idea of increasing consumer demand. The result of this policy is to discourage consumers from saving. Consuming now is better for the economy than saving now and consuming later.

GDP has been around as a concept since the early 20th century. Measuring GDP has been revised many times. While the revisions are driven by economists and technology improvements, the initial construct was political. I’ll have to find the link to a book describing the history of the construction of GDP.

Earlier this summer, the BEA made an announcement they were commencing the production of a new economy wide measure called Gross Output (GO). This measure would also be produced quarterly and follow a revision schedule somewhat like GDP. The eye-opening concept of GO is to measure the output of the economy, or the production of final goods. To use the same terminology as GDP, GO measures the supply within the economy.

I’ll begin to spend some time looking at the differences between the two measures. They are numerous. The categories do not line-up, aggregate growth rates are different, sub-totals are vastly different. This new measure provides a look into the mirror view of the economy and has been lacking in the discussion of how to grow the economy. For now, GO only goes back to 2005, but that is enough data to look at recent changes in the economy and project the near future.

For example, mining is the fastest growing category in the year over year period ending 2014Q2 at 16.1%. That might be a surprise until you understand that mining includes oil and gas production (since it is drilled out of the ground, it is considered mining). Most categories are growing modestly, but mining has been a very strong force within the economy since the beginning of the series.

If GDP is used as a policy directive, then GO should also be expected to be used similarly. Thus, if the political part of the country wants to grow the economy, focusing on the fastest growing part would be a reasonable place to start. Alternatively, you might want to consider growing the largest part of the economy. Mining is only 2.3% of GO, but Nondurable Goods Manufacturing and Real Estate are each 10.1% of the total. Real Estate is problematic as the prices in this category are increasing faster than income and have for several years. Since the physical amount of real estate is fixed, to affect the growth of real estate is only to affect the price of real estate. One result of that policy would be an increase in inflation. Meanwhile, Nondurable Goods Manufacturing is growing at 3.3% year over year. This is less than the overall total of 4.6%. Diving into what constitutes Nondurable Goods and what kinds of policies would impact that part of the economy will be the subject of future posts.

It has been months since I posted. In that period I have come to realize free time is a distant memory. Two jobs are really one too many. In addition, I have been learning some new web technologies. We’ll see if I can incorporate them into something useful here. I have also been getting pressure from the University of Washington to pursue a doctorate degree. After a few months of consideration, that seems impossible due to existing time constraints. While it is still something I am considering, I would need to focus on a particular program, find an institution, and drop one of my jobs. Is it possible to work through a doctorate program with two jobs? I really don’t see how even if I do have the interest. In the meantime, I will continue to practice my writing and analysis skills through this blog.

August 2014 CPI

The Bureau of Labor Statistics released the August 2014 CPI figures today. The ten-year look is on the Blogspot site. What I want to look today is the weightings used in the index. All weightings are set after the December numbers come in, but this will provide a small measure of expectations for changes.


The weighting on Food has been trending down. That isn’t a surprise since the CPI from August 2012 to August 2013 was below the average. For the period of August 2013 to August 2014, the CPI for Food is 2.70%. That is above the average of 1.70% and significantly above the previous year’s figure of 1.39%. The weighting for Food looks to be increasing.


Unsurprisingly, the weighting for Housing has been increasing. The CPI for the period of August 2012 through August 2013 was 2.17%, above the average level. For the period of August 2013 to August 2014, the CPI for Housing was 2.63%, like Food it is above the average and above the previous period’s rate. Even though, the weighting on Housing has been increasing, it looks like that trend will continue.


The weighting on Transportation has been decreasing. With a CPI for the period of August 2012 through August 2013 of 0.05%, that might not be a surprise. For the period of August 2013 through August 2014, the CPI for Transportation is -0.43%, which does seem like a surprise. Transportation looks like it will continue to see a reduction in weighting.

Okay, those are the big three because they add up to almost 73% of the total weight. There are two more areas of interest, though.

Medical Care

The weighting on Medical Care has been increasing. The CPI for Medical Care for the period of August 2012 through August 2013 was 2.34%, well above the average. This area becomes interesting when the CPI for the August 2013 through August 2014 period comes in at 2.09%. While this is still above the average, it is below the previous year’s rate. This part of the CPI will be interesting to watch during the balance of the year since the weighting might remain nearly the same due to the decreasing trend. That could cause some controversy since the rate is still above average.

Education & Communication

The trend for Education and Communication has been up. Given stories of the cost of Education, that isn’t a surprise. For the period of August 2012 through August 2013, the CPI was 1.55%, just slightly above average. For the period of August 2013 through August 2014, the CPI is 1.48%. Not only is that below the previous year’s rate, it is also below the average. If that continues for the remainder of the year, I would expect the weighting to decrease from it’s current amount of 7.087. That might also cause some controversy.

I have wondered why Education and Communication have been grouped together. Perhaps they were just pulled out of Other. Unfortunately, for the past decade those two components have been moving in opposite directions with Education rising and Communication falling. To be clear, Communication has seen a significant amount of deflation. Placing these two items together causes the increase in Education to be muted and difficult to compute.

Still, I have the data:

Sub-Category August 2013 to 2014
Education 3.28%
Books and Supplies 5.18%
Tuition 3.15%
Information Technology -1.18%
Personal Computers -6.85%

Mariners June 2014

Table 1. Batters in June
Player Runs Created RC/Game
Seager 20.1 7.3
Cano 17.2 6.5
Miller 12.8 6.0
Zunino 10.7 4.6

The Mariners end the month of June as a playoff team. They have a 1.5 game lead in the wild card standings and yet are third in the American League West. They have a better record than Baltimore, New York, and Kansas City. The team went 16-10 in the month and that was good enough to get them into this position at the halfway mark of the season. I should say quietly get them into position because there hasn’t been much talk about this team being a playoff team. That will change as we get into July.

I will start off the June recap with the batters again. Robinson Cano and Kyle Seager are on pace to get at least 99 runs created for the season. It has been an extraordinary long time since the Mariners have had two players reach triple digits (maybe 2002?). That will be something to watch. Cano has been performing at this pace all season, but Seager had a great June to push himself to this level. I was opposed to moving him into the four spot of the lineup, but he responded well and the combination of Cano batting third and Seager fourth has created an impact in the middle of the lineup. As a team, there were 119 runs created, or 4.6 per game, during the month. That’s the best on a per game basis for the first three months of the season. They did that while the opposing game score average rose to 53.4 during June. That would lead me to believe they scored often against the opposing bullpen.

One of the measures I am watching is record based on runs scored. The past two seasons featured a break point of 4 runs scored per game before the team reached a 500 record. This year it is 5 runs per game. When scoring five or more runs per game, the Mariners are 29-3. Scoring four or fewer yields a record of 15-35.

Table 2. Starting Pitching
Pitcher GS GSC Avg Ks
Hernandez 6 74.2 54
Young 5 54.2 17
Iwakuma 5 52.0 26
Elias 5 51.4 22
Ramirez 5 50.8 18

That seems like a good transition to the starting pitching. We’ve seen the pressure put on the staff when they need to hold the opponent to three runs or less in order for the team to have a chance to win. The team featured a staff of exactly five pitchers during the month. That is a sign of effectiveness or simply lack of injuries. For June, it was the former where Table 2 shows all five pitchers averaged above a 50 game score. The staff averaged 57.2, a remarkable month. Any discussion about the month has to start with Felix Hernandez. He threw 44.1 innings in his six starts or just over 7.1 innings per start. An average game score of 74.2 is incredible and brought his average for the season up to 65.7. How much of a difference did this make? All four of the other starting pitchers performed worse in June than in May.

If there was an item of concern, it would be innings pitched by the staff. Ramirez threw 23 innings over five starts and Young threw 28 innings over his 5 starts. For pitchers not named Hernandez, there were 20 starts and 113 innings thrown. That’s about 5.2 IP per start. That really puts the test to the bullpen.

Table 3. Relief Pitching
Situation ERA K/9 K/BB IP
High LI 3.00 10.2 4.3 30.0
Low LI 0.62 8.5 4.6 43.2

In looking at Table 3, it appears the bullpen did well during the month. In particular, look at the low leverage index situations. They occur in two types — one where the Mariners are behind and one where they are far ahead. The conclusion, I come to is when the Mariners are behind, the bullpen didn’t let it get worse and when the Mariners are far ahead, they shut down the opposition. Giving the team a chance to come back seems important though I really don’t think this team has the track record of late inning heroics. Nevertheless, the bullpen had a terrific month and those strikeout to walk ratios for the month are something to behold.

Mariners May 2014

Table 1. Batters in May
Player Runs Created RC/Game
Cano 19.6 6.8
Saunders 15.4 6.2
Seager 17.4 5.5
Jones 11.8 4.6
Zunino 9.6 3.6
Smoak 12.0 3.2

When I heard that Robinson Cano was named Mariners’ player of the month for May, I decided I should lead off with the batters. Table 1 includes the four players who created more than 4.5 runs per game plus two more were were above 10.0 runs created or close. Cano created his 19.6 runs in 73 outs, third most on the roster. Smoak created his 12.0 runs and used 95 outs. While I recently wrote a post exclusively about James Jones, maybe I spoke too soon. He stumbled the last week of the month and slipped to only 4.6 runs created per game. He used 65 outs to do that. Maybe I should have written about Michael Saunders who only used 63 outs to create 15.4 runs. I know he doesn’t play everyday, but he made good use of the opportunities he had in May.

Table 2. Starting Pitching
Pitcher GS GSc Avg Ks
Iwakuma 6 60.0 46
Hernandez 6 58.8 38
Elias 7 57.0 52
Young 7 56.6 53
Maurer 5 34.8 44

What I had intended to lead off with was the starting pitching. The top four had great performances in the month if only by the average game score. With 50 being average and 60 starting talk about Cy Young awards, I am sure every Mariners’ fan was very pleased to see Iwakuma back and pitching like there never was an injury. Elias threw a 51.4 average in April and really stepped it up in May. Young only had three starts in April, but a 49.0 average didn’t prepare us for a sharp improvement to 56.6 in May. Unfortunately, Brandon Maurer regressed in May and frequently was pulled before he completed five innings. The potential is there with 44 strikeouts in five starts, but simply too many base runners and runs allowed got him sent back to Tacoma.

The strikeouts column is new this month. The Cleveland Indians broadcast made a big deal about Corey Kluber throwing 60 strikeouts in the month of May. It has been awhile since an Indians pitcher did that. We have King Felix. While he collected 67 strikeouts in April, 38 for May does not seem right. Well, we should at least be pleased to see that Young and Elias got close, though there is little chance either could get eight starts in a month.

Table 3. Relief Pitching
Situation ERA K/9 K/BB IP
High LI 3.62 10.9 2.2 27.1
Low LI 1.71 8.7 3.5 47.1

The bullpen got a little break during May as a consequence of the starting pitching doing so well. They actually were required to pitch four fewer innings in May yet there are eight more games than April. The performances in aggregate were better as well. In both high leverage and low leverage situations, ERA, K/9, and K/BB improved in May over April.

The record for the month was 18-14, an improvement over April’s 10-14. June is not going to be a typical month. The month opens with a make-up game in New York followed by two in Atlanta. The first homestand of the month includes Texas, the only American League West opponent for the month. Immediately after Texas is four with San Diego, split by venue. Maybe James Paxton returns during June. Maybe Taijuan Walker returns during June. It doesn’t seem like we need more starting pitching, but both would be welcome sights in Mariners’ uniforms.

10-Year Yield Channel Break

The 10-year yield broke out of its channel this week. The current channel range of 2.47 to 3.03 has been in place for 341 days. This week saw three closes below the 2.47 lower limit. Time will tell if this signals a change, but a few economists I follow have been making calls for lower yields in the short-term.

Washington State Unemployment

I really enjoy the graphics in the Sunday edition of the Business section of the Seattle Times. The TimesWatch section puts together quite a bit of data into twenty or so graphs that show the state of the state economy alongside the national economy. In the May 25th edition, the graph depicting state unemployment at 6.1% caught my attention. I wondered if the state economy was experiencing the same decreased employment percentage as the national economy is.

For those who do not know, I like to follow the employed percentage of population at the national level. While the economy appears to be recovering if you look at the 6.3% level of unemployment, there is a different picture if you follow the employed percent of population. This measure takes the sum of the employed and unemployed members of the population and divides it by the population of noninstitutional people. Now, while the employed number doesn’t concern itself with full or part-time work, and the unemployed part means you do not have a job but are actively looking for one, the ratio has some suspicious assumptions. Those assumptions are consistent and the best of what we have right now.

For the state of Washington, the same data exists using the same assumptions. In fact, the same Bureau of Labor Statistics web site contains data on each of the fifty states. I gathered the publicly released data on the unemployment rate, the noninstitutional population, the employed, and the unemployed counts for my own analysis. The result is contained on this page, but I will summarize the results here.

The monthly data goes back to 1976. The economy of Seattle was quite different back then as Boeing ruled the employment spectrum. Over the years, Boeing’s influence has waned as the technology sector started to become a much larger employer. I wonder if bio-technology is next, but I think I am getting away from the main point.

The unemployment rate has been on a roller coaster over the years and hit a recent peak of 10% in March 2010. Since then it has moved lower until it reached its current point of 6.1%. Meanwhile, the employed percentage of population looks nothing like the unemployment rate. You can see the end of the boom just as the year 2000 was beginning, but the line didn’t drop below its long-term average of 66.6%. That is remarkable to me and while I suspect it would tell some of the resilience of the state’s economy, I would be more inclined to say it was due to the variety of labor sources in the region.

Where the two line graphs coincide is at the end of their data streams. The employed percentage of population peaked in March 2009. The unemployment rate peaked in March 2010. These line graphs should be negatively correlated — when the unemployment rate is falling, the employed percentage of population should be rising. Instead, the employed percentage of population has fallen from 68.6% to the current level of 63.3% while the unemployment rate has fallen from 10.2% to its current level of 6.1%.

What I have done next is to take the long-term average of the employed percentage of population (66.6%) and applied it to the count of noninstitutional population and subtracted the current employment number. Finally, I divided the result by the noninstitutional population to derive an adjusted unemployment rate, which is currently at 7.2%. The most recent peak was 7.8% in December 2013. In March of 2010, when the published unemployment rate was at its recent peak of 10.2%, the adjusted unemployment rate was 6.0%. Why was it lower? Because the employed percentage of population was 67.5% back then, above its long-term average.

Why do I go to all of this trouble? The difference between the published unemployment rate and the adjusted unemployment rate may not seem like much, but it is 395,000 people. Adding that to the payrolls of Washington state employers means more income, more consumer spending, and more tax receipts for the state. The lower economic activity in the state, as shown by the lower than average employed percent of population, does have an effect. It is something to be concerned about and it is something that should be generating interest.

James Jones

It’s about time that I gave James Jones his own post. As lead-off hitter for the Seattle Mariners, he had done remarkable work having broken through 10 runs created. He has done this in only 57 at bats. This works out to 6.6 runs created per game. To put that in perspective, Robinson Cano has 6.4 runs created per game. Everyone else who has played centerfield this year, in 133 at bats has created 9.6 runs.

At some point, word will get around the league, Jones will slump and he will need to make some adjustments. That could happen next week when the team travels to Tampa Bay, a team that has already seen him and allowed him to get 6 total bases in twelve at bats. For now, it will be nice to see the Mariners with a good lead-off hitter.

National Highway Trust Fund

I was alerted to a news article from Reuters this week on the state of the bridges in the United States. The article is mostly about an analysis by the American Road and Transportation Builders Association, a federation and lobbying group whose stated goal is to “aggressively grow and protect transportation infrastructure investment”. The curious part of the article is about the National Highway Trust Fund. This fund provides monies to states for road and bridge projects.

What is curious to me is the trust fund gets its income from taxes on gasoline and diesel sales. I had read a few years ago that gasoline sales were declining and decided I needed to update my knowledge.

The first stop was the U.S. Energy Information Administration, a government agency that holds data on retail sales of gasoline. The next stop was the Federal Highway Administration of the U.S. Department of Transportation. The result was a new post on The Analytical Road Data.

While I agree that gasoline sales have plunged since the economy went into recession in 2007, it does not seem to have had an effect on the National Highway Trust Fund balance. I broke down the cash flow statement of the trust fund further and noticed that tax receipts are stable. Then I looked at income from all sources and subtracted outlays (what I call Net Income in the graphic) and again noticed the stability of the data series.

The Reuters article mentions the National Highway Trust Fund will run out of funds at the end of this fiscal year (end of August 2014), but it does not mention why. Given the stability of tax receipts and the well managed nature of the expenditures, it doesn’t seem like the fund is in the danger brought forward by the Association.

Before I move away from this topic, the disconnect between the drop in gasoline sales and the tax receipts by the fund cannot go without further analysis. I’ll have to dive into the numbers of the fund to see exactly what the source is since it does not appear to be based on retail sales of gasoline.

Mariners April 2014

Table 1. April Summary
Series Record Runs Scored
per Game
1 4 – 2 5.7 Away
2 2 – 3 2.4 Home
3 1 – 6 3.0 Away
4 3 – 3 4.0 Home

Characterizing the first month of the 2014 season for the Seattle Mariners seems to revolve around the concept of this team still trying to find its identity. With no long winning streaks but one significant losing streak, the team quickly forgot the great start to the season and finished the month below 500. Scoring 4 runs per game still seems to be the magic number for this team as it has been for the past few seasons. It is something I will continue to focus on throughout the season.

Table 2. Starting Pitching
Pitcher Games Started Average Game Score
Hernandez 6 64.0
Elias 5 51.4
Ramirez 5 43.0
Young 3 49.0
Paxton 2 65.5
Maurer 2 43.0
Beavan 1 43.0

King Felix arrived with a great beginning to the season. With continued talk of reduced velocity, his strikeout totals were up over the past few seasons and he had a great start to the 2014 season. James Paxton also got off to a good start but got sidelined with an injury. As long as Paxton was rolling, the loss of Iwakuma wasn’t hurting the team. But after only two starts, the team was dipping deep into the well to cover some games. Roenis Elias was a surprise at how well he opened the season. Having signed with the thought of being a fourth starter at best, he stepped up and had a good April.Looking forward, Iwakuma will be back in May. Paxton and Walker might be back by the end of the month, though that might be too optimistic. Hernandez, Iwakuma, and Elias are probably league average for a one through three starting staff. I expect the starting pitching to be fine in May.

Table 3. Batting
Batter Runs Created Runs Created
per Game
Cano 13.55 5.4
Seager 12.20 5.1
Smoak 10.77 4.4
Hart 10.67 5.0
Zunino 8.30 4.1
Ackley 7.67 3.4
Almonte 7.50 2.5
Miller 5.91 2.1
Saunders 3.55 2.8

As expected, Robinson Cano tops the runs created board. Kyle Seager was performing very poorly until the last series of the month when he co-won the American League Player of the Week award. Corey Hart is also producing well and hasn’t let Safeco Field intimidate him. After the top three, Justin Smoak seems to be contributing occasionally but not consistently. That means no change from him. Mike Zunino has shown improvement over last year and seems to be handling the catching role well. After the top five, there are issues. Dustin Ackley continues to have long stretches of no contribution. Abraham Almonte is not doing well as a lead-off hitter. At one point in the final week of April, he was tied for second in the league for strikeouts. Brad Miller has not continued his success of 2013 into this year and is no longer considered an everyday starter. Michael Saunders is playing part-time and just barely makes this list. Stefan Romero just missed the top nine and might get increased playing time. He doesn’t produce much but his consistency is better than others on this list. Run production by the team is still below league average. That will not change until there are more than three players creating more than five runs per game. I’ll be watching for the continued improvement of Zunino and the return of Miller. In addition, the team needs to find a better lead-off hitter.

Table 4. Batting Order
Order Runs Created
per Game
1 3.18
2 1.60
3 5.40
4 4.61
5 2.85
6 6.53
7 2.49
8 4.59
9 2.02

Just to emphasize the Mariners’ struggles at the top of the batting order, I added Table 4 which shows the Runs Created per Game by batting order. What I didn’t expect to find was the problem in the number two slot. This really solidifies the trouble the Mariners will have scoring since their top two spots in the batting order cannot seem to create runs. Increased run production should be a goal and altering the top two spots should be the top target.

Table 5. Fielding Position
Position Runs Created
per Game
DH 3.11
1 0.00
2 3.71
3 4.47
4 4.94
5 4.67
6 2.06
7 2.85
8 2.48
9 4.15

I don’t recall that I published Runs Created by Fielding Position in 2013. It is something that I remember watching, but a different coaching staff and a different roster make this year stand out. In particular, the outfield is something to behold and not in a good way. While right field is creating just over the magic number of four, left and center are dreadful. The strength of the team is in the infield, but the outfield isn’t helping. This needs to turn around if the team wants to reach 500 for the season. It is humbling to focus on a 500 record as a goal, but it can be achieved if the outfield starts producing.

Table 6. Relief Pitching
Pitcher IR IS High LI Low LI
Farquhar 5 1 5 7
Furbush 8 4 8 4
Medina 5 1 7 5
Wilhelmsen 4 2 4 7
Beimel 5 0 3 7
Rodney 1 0 7 3

Table 6 shows part of the current relief staff. I only included the pitchers with the most appearances. The acronyms are: IR — inherited runners; IS — inherited runners who scored; and, LI — leverage index. On the plus side, Joe Beimel has not allowed an inherited runner to score. In his ten appearances, seven were in low leverage situations. I would recommend he be brought into more high leverage situations. In fact, he might simply swap with Charlie Furbush who has allowed half of his inherited runners to score. In his eleven appearances, eight were in high leverage situations. Furbush may need to have his role altered until he can consistently get outs. I’m not sure what role Danny Farquhar is playing. He was the closer at the end of the 2013 season but is being used in more low leverage situations than high leverage situations so far. It seems that he too should have a more prominent role in the bullpen.

Table 7. Leverage Results
High 4.50 1.563 9.3 1.4 32.0
Low 2.72 1.273 7.4 2.1 46.1

I like looking at the bullpen results by leverage index to see how well they are performing in important situations. For Table 7, there is a very notable difference in outcomes between low leverage situations and high leverage situations. While I applaud the bullpen for their low ERA in low leverage situations, this isn’t a team that has created an offense which can come from behind. In high leverage situations, the bullpen is allowing too many baserunners, as seen in the WHIP, and consequently allowing too many runners to score, as seen in the ERA. An alarming indicator is the K/BB ratio in high leverage situations. There have been 24 walks in 32 innings pitched. Time to make a change. I would start with what I mentioned in the preceding paragraph. Two small changes could have a helpful impact and not allow the team to lose games in the late innings.

For the month of April, the team was outscored 99 to 91 and their record was 10 – 14. The opposition had a better aggregate game score and a higher number of runs created. I’m looking forward to a few subtle changes in May. The bigger changes that are needed may need to come from a few players using the role they have been given to produce more.

Monetary Policy and Financial Exuberance

This note came in from Seeking Alpha’s Wall Street Breakfast:

The Chicago Fed’s Charles Evans doesn’t expect the first rate hike until the 2nd half of 2015 – somewhat later than Yellen’s “six-month” (from QE’s end) remark which suggested a boost as soon as April 2015. Speaking to reporters after his speech, Evans suggested holding off on hikes until 2016 could be appropriate given the state of the economy. Nevertheless, Evans sees a Fed Funds rate of 1.25% by the end of 2016 – the low end of FOMC guesses, but 25 basis points higher than his forecast three months ago. As for hiking rates to cool “financial exuberance” – an idea seeming to gain a little traction with some Fed members in recent days – Evans says “monetary policy is not the best tool to mitigate this risk.”

From my perspective leading up to 2007, monetary policy was the cause of financial exuberance. It seems to me it should also be part of the solution. Is it the best tool to mitigate the risk? Perhaps not, but of the remaining tools (fiscal policy, taxes, and regulation), it is the only one available right now to be used.There doesn’t seem to be any resolve in the chambers of Congress to even acknowledge this risk exists.

Sometimes waiting for the best tool to become available starts to look like indifference. This could lead us to the same outcome as we saw in 2008. I do not condone this level of abdication of responsibility. I especially dislike it when those who perpetrate the situation tell us the problem now is ours to resolve.


Get every new post delivered to your Inbox.