Ep. 11: “Clickbait Works! The secret to getting views with the YouTube algorithm” by Linus Wilson on the Finance Professor Podcast

Professor Linus Wilson discusses his new paper “Clickbait Works! The secret to getting views with the YouTube algorithm on episode 11 of the Finance Professor Podcast.” There is a lot of contradictory advice about what metrics the largest video sharing site in the world and the second largest social network promotes. Using a new data set available to YouTube creators starting in 2018, Dr. Wilson finds that click-through rates are by far the most important predictor of a new video getting views from YouTube’s black-box recommendation system.

The link is https://ssrn.com/abstract=3369353

Click the download button for a free download. Next, if you don’t want to register or login with SSRN, click “Download without registration” under the gray text under the picture of Greg Gordon.

“Clickbait Works! The secret to getting views with the YouTube algorithm”

By Dr. Linus Wilson


In 2018, YouTube began releasing click-through rates (CTR) data to its video creators. Since 2012, YouTube has emphasized how it favors watch time over clicks in its recommendations to viewers. This is the first academic study employing that data to test what matters more for views on YouTube. Is watch time or CTR more important to getting views on YouTube? This paper finds no to limited evidence that higher percent audience retention or and total average watch time per view are associated with more views on YouTube. Instead, videos with higher CTR got significantly more views as did videos on trending or newsworthy topics. The marginal benefit in terms of views scaled by subscribers of increased CTR is between 71 and 318 times larger than the marginal benefits of increased watch time per view.

Journal of Economic Literature Codes:  D12, D22, D26, D83, D85, L15, L21, L82, L86, M15

Keywords:  YouTube, algorithm, search, discovery, video, CTR, click-through rates, clickbait, watch time, audience retention, neural networks, recommendation systems

Wilson, Linus, Clickbait Works! The secret to getting views with the YouTube algorithm (April 9, 2019). Available at SSRN: https://ssrn.com/abstract=3369353

For all of Linus Wilson’s research go to



Secret Wall Street Bailout Uncovered – Broken Bucks: Money Funds & Taxpayer Guarantees

Find out about the secret $2.7 trillion bailout of Money Market Mutual Funds MMMFs in 2008. The collapse of Lehman Brothers its commercial paper default caused the Primary Reserve Fund to “break the buck” or sell for less than $1.00 per share. Linus Wilson presents his paper at the Southwest Finance Association (SWFA) & Federation of Business Disciplines (FBD) conference at the Hyatt Regency in Houston, Texas on March 15, 2019.

Wilson, Linus, Broken Bucks: Money Funds that Took Taxpayer Guarantees in 2008 (August 28, 2015). Available at SSRN: https://ssrn.com/abstract=2195358 or http://dx.doi.org/10.2139/ssrn.2195358

The U.S. Treasury rolled out a bailout guarantee on September 19, 2008 without Congressional approval using the exchange rate stabilization fund led by Hank Paulson, David Nason, and Steve Shafran. Paulson and Shafran were Goldman Sach alums (p. 263) ON THE BRINK by Henry Paulson.

Broken Bucks: Money Funds that Took Taxpayer Guarantees in 2008
42 Pages Posted: 2 Jan 2013 Last revised: 29 Aug 2015
Linus Wilson
University of Louisiana at Lafayette – College of Business Administration

Date Written: August 28, 2015


This is the first study to look at the characteristics of funds accepting the $2.7 trillion taxpayer guarantee of money market mutual funds during the 2008 financial crisis. Funds with lower asset maturities were significantly less likely to need federal or sponsor bailouts. Fund shares that benefited from Federal Reserve’s asset-backed commercial paper program were significantly more likely to get bailed out by taxpayers and sponsors. Finally, the paper tests if funds adhering to the SEC’s 2010 liquidity reforms prior to their enactment were less likely to be bailed out in 2008.

Keywords: breaking the buck, bailout, Dodd-Frank, DLA, exchange rate stabilization fund, Financial Stability Oversight Council (FSOC), guarantees, liquidity, money market mutual funds, Primary Reserve Fund, regulation, SEC, Securities and Exchange Commission, U.S. Treasury, WAL, WAM, WLA

JEL Classification: G01, G18, G22, G23, G28, H12, H81, L5

Music by http://www.BenSound.com
(c) Linus Wilson, 2019

U.S. Treasury portrait of Secretary Henry Paulson was completed in 2010 image at https://www.treasury.gov/about/history/pages/Steven Shafranhmpaulson.aspx
by Aaron Shikler

Ranking the Fed Doves & Hawks | Yellen to Martin FOMCs by Linus Wilson

This video ranks the Federal Reserves based on the tenure of their chairs from William McChesney Martin, Jr. to Janet L. Yellen, using data from 1958 through 2018.

This reading of “A Dove to Hawk Ranking of the Martin to Yellen Federal Reserves” by Linus Wilson.

Inflation “doves” are willing to tolerate more inflation than inflation “hawks.” Comparing the Taylor (1993) rule and core inflation to the effective fed funds rates, it is found that the Yellen Fed is the most dovish Fed since 1958.

The tenures of the following Federal Reserve (FOMC) Fed Open Market Committees are analyzed based on core inflation (CPI-U) without food or energy prices, unemployment, fed funds rates and the Taylor rule:
Janet L. Yellen
Ben S. Bernanke
Arthur F. Burns
G. William Miller
William McChesney Martin, Jr.
Alan Greenspan
Paul A. Volcker

The paper quotes or cites some of the speeches of Janet Yellen.

It was written by Linus Wilson, Associate Professor of Finance at the University of Louisiana at Lafayette. The views expressed are his alone.

The Yellen Fed is found to be the most dovish in history based on its setting of short-term interest rates relative to inflation. This paper looks at the interest rate setting policy of the Federal Reserve going back to the chairmanship of William McChesney Martin, Jr. and ending with the Janet Yellen’s tenure as chair. The Yellen Fed lacked a recession or banking crisis that may have justified the negative real interest rate policy of the Bernanke Fed. For its four years, the Yellen Fed succeeded in having falling unemployment and low inflation with negative real interest rates.

The link to get the paper is below:
Wilson, Linus, A Dove to Hawk Ranking of the Martin to Yellen Federal Reserves Available at SSRN: https://ssrn.com/abstract=2529195 or http://dx.doi.org/10.2139/ssrn.2529195

The views expressed are of Linus Wilson alone.
Music by http://www.BenSound.com
Public domain photos of the Fed chairs from the U.S. Federal Reserve.
(c) 2018, Linus WilsonDoveThumb1280by720

Find the Cost of DEBT with free FINRA data

In this video, Professor Linus Wilson calculates the weighted average cost of debt using FINRA data. Calculating the cost of debt is important for corporations making capital budgeting decisions as a key component of the company cost of capital as measured by the Weighted Average Cost of Capital (WACC). Finding a company’s cost of debt is important for business valuation.

FINRA stands for the Financial Industry Regulatory Authority.


Music by http://www.BenSound.com

We mentioned chapter 14 (Cost of Capital) in

Stephen A. Ross, Randolph W. Westerfield, and Bradford D. Jordan, 2013, Fundamentals of Corporate Finance: Alternate Edition, 10th ed., New York: McGraw-Hill Irwin, (ISBN: 978-0-07-747945-9).

This video does not represent the views of the University of Louisiana at Lafayette, FINRA, or the authors of Fundamentals of Corporate Finance. This is not investment advice.

Take the 25-mile “Caravan” Challenge

I don’t understand the hype and fear associated with the “Caravan” of Hondurans walking over 1,000 miles across Mexico. I am still fascinated by it as an endurance challenge. I believe every American would change their life (for the better) if they would try to live one part of one day in these refugees footsteps. The CNN report that I watched said they were going to walk 25-miles that day with over 1,000 miles to go to the USA border. (They had 40 more days of walking ahead!)

This is my proposal as a 3-time marathon (26.2 mile) finisher and one-time marathon failure. This is my proposal as someone who has sailed his boat at an average speed of less than 5 mph over twenty percent of the earth’s circumference. This is my proposal as one who has read the stories of Pacific Crest Trail (PCT) and Appalacian Trail through hikers and the ones who could not complete those challenges:

Walk for 25-miles in one day.

Take your money, cell phone, water, and any other modern convenience that you have, walk on the sidewalks of your home town or its parks. Walk in a straight line or in a loop. Stop for meals and drinks, but try to walk the distance. Perhaps you will cover 2, 8, 10, 20, 25, 26.2, or 30 miles before succumbing to tiredness or injury. Then, get a ride to your comfortable climate controlled bed (which the Honduran walkers lack) and see if you don’t admire them. After walking 25-miles in one day, see if you don’t believe they are superheroes on par with Rocketman or Wolverine in the comic books.IMG_2351

The author hiking in the island of Ua Pou, Marquesas, French Polynesia in June 2017.

The Fed Bailout of CMBS: Toxic Asset Subsidies & the Early Redemption of TALF Loans | MEA Conference

Nonrecourse loans can lead to option payoffs for the borrower. The securitzation market got a boost from the Federal Reserve during the financial crisis. Dr. Linus Wilson, Associate Professor of Finance at the University of Louisiana at Lafayette presents and reads his working paper:


“Toxic Asset Subsidies and the Early Redemption of TALF Loans”

This paper develops a formula to numerically estimate the unsubsidized, fair-market value of the toxic assets purchased with Federal Reserve loans.  It finds that subsidy rates on these loans were on average 33.9 percent at origination.  In contrast, by the 3rd quarter of the 2010, there was on average no subsidy in TALF loans.  The theoretical model is used to predict the early redemption of Term Asset-Backed Securities Loan Facility (TALF) loans used to purchase commercial mortgage backed securities (CMBS).  The predictions of the model are strongly supported by the data.  In addition, this paper looks at the determinants of early redemption.  CMBS originated inside the peak bubble years of 2005-2007 were much less likely to be redeemed early.  The giant investment managers, Blackrock and PIMCO, were much more likely to redeem their TALF loans early than smaller investment managers.

Journal of Economic Literature Codes:  G12, G13, G18, G21, G28, G38

Keywords:  alternative investments; bailout; banking; Blackrock; call options; commercial mortgage backed securities; CMBS; CDOs; Dodd-Frank Financial Reform Law of 2010; emergency lending; EESA; Emergency Economic Stabilization Act; lending; Legacy Securities Program; mortgages; PIMCO; Public-Private Investment Partnership; PPIP; put options; TALF; Term Asset-Backed Securities Loan; TARP; Troubled Asset Relief Program; toxic assets

Come and see Dr. Wilson present the paper at the 2018 Midwest Economics Association meeting on March 24, 2018.

The MEA conference in at the Hilton Orrington in Evanston, Illinois.


This paper was read on the Finance Professor Podcast episode 8 which is at


Music by http://www.BenSound.com

This paper does not represent the views of the University of Louisiana at Lafayette or the Midwest Economics Association.




Will “rock star” academics’ video lectures replace live professors?

Distance learning begs the question of whether technology will lead to a great consolidation of the academic ranks as recorded lectures of “rock star” academics displace your local in-person lecturer. My little experiment leads me to think that there is no replacing an in-person lecture. If students’ learning is a priority, make them come to class and don’t assume a video lecture is a substitute.

Class cancellations in the spring 2018 term due to an ice storm meant that one more lecture of my international finance class was cancelled in section 2 versus section 1. This meant that I could record the section 1 lecture on Thursday and post a recording of it before the next Tuesday’s class in section 2. My university allows up to 49% of lectures in non-distance learning classes to be virtual. Up until this time, I had conducted 100% of my lectures as live with no video recording at the University of Louisiana at Lafayette.

I told my section 2 class that they were required to view the video all the way through and take notes from a link that I posted on the class-management, Moodle page only accessible to students. The video was made unlisted so that it did not show up in search and suggested features in YouTube. That link was clicked 31 times out of a class of 35 students. Some students likely clicked the link multiple times, and several did not bother to click it at all. The average watch time on the 75 minute video was 10 minutes or about 13%. Compare that to a typical lecture leading up to the virtual where 31 out of 35 students attend and are in their seats for over 95% of the lecture. Obviously, the in-person lecture is going to lead to better learning outcomes.

It did seem to on the test. I curved the video lecture’s test based on their lower scores so the curved results in the video lecture section and the regular lecture sections curved scores are similar. The raw, uncurved average in the video lecture section was 66% and was 74%. The raw medians in the video and in-person lecture section was 69% and 75% respectively. There were 5 lectures covering the test in question. For the video section, 4-out-of-5 lectures were in-person lectures. For the other section, 5-out-of-5 lectures were in-person. There is every indication that, if I held more video lectures, the video section’s scores would have been abysmal. I do give different test questions in each section and some of the variation in test scores could reflect the different questions, but the question topics were almost identical. The question topics should be entirely predictable to a student taking notes in the lectures.

I used 3 cameras and edited the video with screen shots of the lecture slides at various points in addition to my working out examples on the board. Nevertheless, I did not edit our the normal pauses mid-lecture that an in-person student would have sat through. For fun, I have an entertainment oriented channel with over 4,000 subscribers on YouTube. Its typically, 15 minute videos with a very focused niche audience only average 50% audience retention. Longer videos typically have lower percentages viewed.

50% audience retention will lead to failing grades in education. That is why in-person lectures and in-person education will dominate video lectures. Since academic “rock stars” can record videos but cannot appear in-person to thousands of sections across thousands of universities, the local professor or lecturer will be much more effective in educating students than on-demand, video education.

Social pressure keeps most students from walking out mid-lecture. There is no such social pressure when it comes to watching a video. Other parts of the live experience may also facilitate learning in ways we don’t understand. There is no way that every lecture can be as entertaining multi-million dollar budget Hollywood blockbuster. That is why video education will fail, and in-person education is here to stay.