Guadalajara: International Book Fair: Higher Education for the Future of Work

On 1 December 2019, Gabriela Ramos gave a keynote opening a panel entitled “Higher Education for the new future of work” at the Feria Internacional del Libro (International Book Fair) in Guadalajara, Mexico. The panel included Jaime Valls, Secretario General, Asociación Nacional de Universidades e Instituciones de Educación Superior, Pankaj Mittal, Secretaria General, Asociación de Universidades Indias, Ricardo Villanueva Lomelí, Rector General, Universidad de Guadalajara, Enrique Graue Wiechers, Rector, Universidad Nacional Autónoma de México, Francisco Marmolejo, Líder Especialista de Educación Superior, Banco Mundial, Raúl Beyruti, CEO, GinGroup.

Ladies and Gentlemen,

It is a great pleasure to be with you here today.

I want to thank the University of Guadalajara and Ricardo [Villanueva Lomelí, Rector General, Universidad de Guadalajara] for hosting this event.

Mexico has been a reform champion! The country has launched ambitious reforms in a broad range of areas, and of course that includes education, thanks to all of you here. But let’s keep moving – there is no room for complacency.  

Education holds key to almost all the social and economic problems we face today. Higher education is a key driver of inclusive growth.

And, this topic is very close to my heart, since I spearhead the OECD Inclusive Growth Initiative. In fact, before flying to Guadalajara, I hosted the annual OECD Global Strategy Group meeting in Paris where OECD members and key partners discussed the most topical issue of “ageing”.

Ageing is not just about older people, it makes us think seriously how important it is to take a life-course approach to the problems we face today in education and in the labour market.

How do we unlock the potential of older workers and how to empower them to stay active in the labour market? I believe that the solution lies in education, and I mean starting strong from early childhood education through to higher education.

Indeed, the number of young people with a higher education qualification is expected to surpass 300 million in OECD and G20 countries by 2030, and of course we should welcome this great progress in access to education.

But having more access is simply another  step forward. The most important mission for us is to ensure that the skills acquired by students in higher education could align well with the skills needed in the labour market. The time spent in education will otherwise be wasted.  

What are the challenges in higher education today?

It is all about skills! Challenges are:

  • Anticipating skills needs;
  • Meeting immediate skills needs;
  • And ensuring the skills it produces are effectively used later

Today, we face a high level of skills mismatch.  

  • Employers often complain that they cannot find workers with the required skills.
  • At the same time, large numbers of higher education graduates face difficulties in finding job opportunities matching their qualifications.

This challenge of skills mismatch is common in all OECD countries. And Mexico is no exception!

  • The results of the General Exit Exam for Bachelor’s Degrees (EGEL) confirms that more than half of the students cannot demonstrate the basic level of knowledge and skills required in their field of study by the time they complete their bachelor’s degree. 
    • Out of all subject areas, the highest failure rate could be found in the areas of education sciences. The failure rate increased from 36% to 51% between 2011 and 2016.

We have documented all our analysis in our report Higher Education in Mexico, launched this year.  

  • In Mexico, 44% of bachelor’s graduates are overqualified for their jobs.[1]
  • Generally, students have limited labour market information given to them when they choose their programme.
  • Now, more specifically on Mexico, we lack diversity in the programmes offered.
    • Higher education institutions tend to deliver programmes that are likely to attract high enrolments.
    • They also go for less costly programmes in terms of staff and infrastructure.
    • As a result, nearly half of higher education programmes are offered in social sciences, administration and law, and 71.9% are offered at the bachelor level.
    • On the other hand, the current innovation capacity is very limited in Mexico.
      • There are only 0.7 R&D personnel per 1 000 employees in Mexico, compared to 7.7 in OECD countries, 25% of whom work in business [61% OECD average]. 
      • Only 17% graduated from engineering and 8% from ICT programmes.
      • The OECD Skills for Jobs database identifies shortages in science and engineering professionals and ICT associate professionals.
  • So, the demand for these workers exceeds the supply!

What we need in higher education today is to offer a more diverse range of programmes in different fields of study.

  • Mexico could train master’s and doctorate students to increase R&D activities and drive innovation in the private sector, particularly in its strategic industries (e.g. energy, automobile and aerospace).
  • In addition, to build a solid high-tech entrepreneurship ecosystem, Mexico needs to provide students with entrepreneurial skills so that they can create and grow their own start-ups and eventually employ others.

And let’s not forget that we also need to adapt a more diverse student base and to provide support to students from less advantaged backgrounds to avoid drop-out.

Exactly what skills matter?

I will be launching the 2018 version of OECD PISA results in a few days here in the University of Guadalajara.

This assesses 15 year-old students’ performance in reading, mathes, science  and we are increasingly incorporating digital aspect to our assessment to reflect the changes represented by digitalisation.

The skills they need are not the same skills we (me and you) needed! Then the skills’ provider – educational institution – should also be held accountable to find solutions for the future.

The future demand for skills

Today, we cannot discuss anything without mentioning the huge impact rapid digitalisation has on our lives and economy.

  • And for good reason – digitalisation is transforming the way we work, where we work, what we do at work, as well as the skills needed if you want to remain in employment.
  • And not just whatever jobs, we need quality jobs.
  • Across OECD countries on average, we project that 14% of jobs are at high risk of automation, while a further 32% will likely change significantly due to technological change. 
  • In Mexico, the risk of automation is an urgent concern, particularly as the share of employment in the manufacturing sector is higher than the OECD average (17% vs 14%).

Meeting immediate labour market needs

Higher education institutions continue to face important challenges in meeting immediate labour market needs to ensure a smooth transition into the future of work.

Reaping the benefits of technological change requires managing inevitable disruptions. Our data shows that too many individuals in OECD countries lack even the basic skills to succeed in a technology-rich environment.

  • According to the OECD PIAAC, half of Mexican 16-65 year-olds lack basic literacy (OECD average 19.7%).
  • And 60% of them lack basic numeracy (OECD average 23.5%).
  • These proportions are among the highest observed in the PIAAC participating countries, and are similar to those found in Chile, Ecuador, Peru and Turkey.
  • This is worrisome! Many people are lacking the basic skills to thrive in the digital economy.
  • Even among tertiary-educated adults, only 26% demonstrated good problem-solving skills in a technology-rich environment in Mexico (OECD average 48%).
  • This means that holding a tertiary degree does not always guarantee a high level of skills.
  • Mexico needs to do much more to raise the quality of their education systems.

And as mentioned, we need to ensure that the skills produced by higher education are effectively used. This is the most challenging mission for Mexico as I said in the beginning.

  • Finding a good job can be more difficult for Mexican higher education graduates than for their peers in other OECD countries.
  • In addition to over-qualification and partly as a result of the skills mismatch, young workers with higher education degrees in Mexico face the challenge of informality.
  • Although the prevalence of informal employment is lower for young higher education graduates (26.7%) than for workers in the same age group who completed only upper secondary education (45.8%), more than one quarter of the most qualified workers in the country have no social security or pension coverage.  
  • At the same time, almost half (45.7%) of young higher education graduates work in a job for which no higher education qualification is required.
  • Over qualification brings the problem of increasing informality for young workers.
    • On average, 14.5% of young higher education graduates do not participate in the labour market [OECD average 10.7%].
    • This places Mexico in a disadvantaged position, as the skills of these graduates are not used.

Last but not least, let me highlight that women continue to be under-utilised. And education should provide answer.

  • In Mexico, although women represent 53.1% of first-time graduates, many women with a higher education degree do not participate in the labour market.
    • Their inactivity rate is three times higher than that of male graduates (21.3% vs 6.9%)
  • And there is a glass ceiling! I met with business women this morning to discuss exactly this!
    • In 2016, only 5.2% of Mexican women had a seat on the boards of the largest publicly listed companies (20% OECD average).
  • Highly skilled women who are not participating to their full capacity in the labour market present a particularly large untapped potential to boost Mexico’s economy.

Against this backdrop, the number of higher education graduates in Mexico is expected to increase, and recruitment for strategic and specialist positions is expected to get even more difficult.

So this is becoming an even more urgent mission for us.

How do we tackle these pressing challenges of skills misalignment with both the future of work and current labour market?

  • Only the joint efforts from stakeholders in higher education, employers, and governments can improve the alignment of skills and knowledge, and thus the contribution of higher education graduates to productivity and economic growth.

The report I highlighted earlier (“Higher Education in Mexico: Labour Market Relevance and Outcomes”) presents a set of recommendations in three key areas where government and higher education institutions can effectively collaborate to equip students with skills relevant for the labour market.

  • First, they must aligning higher education with the changing needs of the labour market.
    • With the rapid changes in skills demanded of workers, they will need more flexibility and the possibility to return to higher education at a later stage of their life.
    • Distance education and online education in Mexico are growing and can be key in this regard.
  • Secondly, we must help  students succeed in higher education and the labour market.
    • This means creating a better pathways into and across higher education and a greater emphasis on lifelong learning .
    • The OECD has created our HEInnovate (Higher Education Innovate) programme which allows higher education institutions to assess the extent to which they encourage and facilitate entrepreneurial and innovative spirit in their students.
    • Additionally, following the example of NiñaSTEM Pueden, which we launched in Mexico in 2017, we can bridge over- or underrepresentation of certain profession with role models. Connecting students with role models in less popular but necessary careers can change perceptions of these paths and encourage more students to enter these fields.
  •  Lastly, we need a whole-of-government approach to enhance labour market relevance and outcomes.

Ladies and Gentlemen,

We invite government authorities, leaders of higher education institutions and employers to continue working together to optimize the quality, equality and relevance of higher education in Mexico.

An first-class higher education system is essential to drive stronger, more inclusive and more sustainable growth.

Thank you very much.

[1] Higher Education in Mexico, OECD, 2019

NiñaSTEM Pueden at Kidzania in Guadalajara

On 2 December 2019, Gabriela Ramos participated in a mentoring event of the NiñaSTEM Pueden initiative at Kidzania in Guadalajara with mentor Carmen Rodriquez Armenta (Dirección de Educación Superior Universitaria representada por Carmen Rodríguez Armenta – Directora General y Mentor de la Iniciativa) and over 100 girls participating.

Mexico: Presentation of PISA 2018 Results

Gracias, Ricardo Villanueva Lomelí, Rector de la U de G.

Me da mucho gusto estar aquí con usted, con el Secretario Moctezuma y con todos ustedes para compartirles los últimos resultados del emblemático Programa de Evaluación Internacional de Estudiantes de la OCDE. Dado que nos acompañan periodistas aquí y a través de un seminario web [con periodistas de ALC], permítanme enfatizar que estos resultados están bajo estricto embargo hasta el lanzamiento oficial mañana (3 de diciembre) a las 9 a.m., hora de París.

Explicación general de PISA 2018

En las últimas dos décadas, esta encuesta internacional trienal se ha convertido en el principal criterio del mundo para evaluar la calidad, la equidad y la eficiencia de los resultados del aprendizaje en los distintos países.

En 2018, 600.000 estudiantes que representan a 32 millones de estudiantes de 79 países y economías realizaron el examen de 2 horas acordado internacionalmente, que contiene preguntas de lectura, ciencias y matemáticas. En la edición 2018 de PISA participaron 10 países de América Latina (Argentina, Brasil, Chile, Colombia, Costa Rica, República Dominicana, México, Panamá, Perú y Uruguay).

Mañana, la OCDE lanzará 3 volúmenes de PISA 2018.

  • Volumen I: lo que los estudiantes saben y pueden hacer
  • Volumen II: donde todos los estudiantes pueden tener éxito
  • Volumen III: ¿Qué significa la vida escolar para la vida de los estudiantes?

¿Que hay de nuevo en PISA 2018?

  • El Marco: Hemos introducido un Nuevo marco para la prueba de lectura, y esta tiene la finalidad de reflejar los cambios que hemos estado viendo en la educación desde 2009 y, por supuesto, como era de esperarse se refiere a la digitalización.
    • Durante la última década, la lectura se ha movido cada vez más a formatos electrónicos y ha proliferado la cantidad de material disponible para leer. Como resultado, la evaluación de lectura PISA 2018 puso mayor énfasis en la capacidad de encontrar, comparar, contrastar e integrar información a través de múltiples fuentes de texto. Por ejemplo, tenemos …
      • Un nuevo proceso: Implementamos una prueba adaptativa en la evaluación de lectura.
        • Los estudiantes que obtuvieron buenos resultados en las primeras partes de la evaluación de lectura recibieron preguntas más difíciles en la siguiente parte de la evaluación.
        • Por el contrario, los estudiantes que obtuvieron malos resultados en las primeras partes de la evaluación recibieron preguntas más fáciles en la segunda parte de la evaluación.
  • El Contenido: Un cuestionario opcional sobre el bienestar de los estudiantes, fue distribuido en los países por primera vez en 2018.
    • Y más contenidos nuevos están por venir en el future: Tendremos una nueva competencia global que fue medida: la innovación. Sus resultados serán presentados en 2020.
    • Países participantes: 79 países y economías participaron en PISA 2018.
      • Varios países participaron en PISA por primera vez: Bileorusia, Bosnia y Herzegovina, Brunei Darussalam, Marruecos, las Filipinas, Arabia Saudita y Ucrania.

Permítanme presentarles algunos de los resultados clave.

Primero, desempeño en lectura.

Al igual que los ciclos anteriores, la evaluación de 2018 cubrió lectura, matemáticas y ciencias, con el enfoque puesto en la lectura en el entorno digital. [Esta es la primera vez que evalúa la lectura como un dominio principal con evaluación basada en computadora.]

Se puso mayor énfasis en la capacidad de los estudiantes para navegar por textos complejos e integrar información de múltiples fuentes, pues estas son habilidades clave para una participación exitosa en el mercado laboral en estos tiempos de rápida digitalización. Las habilidades digitales serán fundamentales no solo para superar con éxito la educación superior, sino también para la vida social y cívica en el siglo XXI.

Y no olvidemos que PISA juega un papel importante para monitorear el progreso de los Objetivos de Desarrollo Sostenible (ODS) de las Naciones Unidas. El nivel de competencias 2 en PISA se usa para evaluar el progreso del Objetivo 4.1, que mide la proporción de jóvenes con al menos competencias mínimas en lectura y matemáticas.

Ahora, permítanme presentar algunos resultados concretos.

  • Los estudiantes de 15 años de edad en cuatro provincias o municipalidades en China y Singapur destacaron al sobre pasar con un amplio margen a sus pares de todos los otros países participantes en lectura.
  • Por el otro lado, casi 1 de cada 4 (23%) estudiantes en los países de la OCDE tuvieron dificultades con los aspectos más básicos de lectura, tales como identificar la idea principal de un texto de moderada longitud o conectar piezas de información provistas por diferentes fuentes.

¿Y qué pasa con nuestra región, América Latina?

  • 79% de los estudiantes se ubicaron por el nivel mínimo de competencia lectora (Nivel 2) en la República Dominicana. [Chile fue un poco excepcional en la región con solo 32% de sus estudiantes por debajo del Nivel 2.]
  • Los 10 países latinoamericanos obtuvieron puntajes inferiores al promedio de la OCDE [de 487 puntos]. La excepción fue Chile, que obtuvo el puntaje más alto de estos países[452 puntos].
    • El puntaje de Chile en lectura fue tan alto como el de Grecia, Malta y la República Eslovaca. [México tuvo 420 puntos] 
  • Lo que muestra este resultado es que los estudiantes tienen dificultades con aspectos básicos de la lectura, como identificar la idea principal en un texto de longitud moderada o conectar información proporcionada por diferentes fuentes.

Esto es extremadamente alarmante. Sin estas competencias básicas, los estudiantes estarán marginados en la sociedad. No solo eso: los países y las economías no podrán aprovechar plenamente los beneficios de los avances tecnológicos. Y lo más importante: los estudiantes tendrán dificultades en su vida posterior en términos de carrera, ingresos, tipo de profesión y bienestar.

En segundo lugar, las tendencias en el desempeño.

PISA brinda la oportunidad de medir las tendencias en el desempeño de los estudiantes a lo largo del tiempo. Los casos de algunos países demuestran que el rápido progreso del desempeño de los estudiantes es posible a pesar de sus desafíos económicos y sociales.

  • Entre los países de la OCDE, Estonia ha avanzado constantemente hasta la primera posición, a pesar de que su gasto por estudiante sigue siendo alrededor de un 30% más bajo que el promedio de la OCDE.
  • Portugal avanzó al nivel promedio de la OCDE a pesar de haber sido fuertemente afectado por la crisis financiera.
  • Algunos países que aún se desempeñan muy por debajo del promedio de la OCDE vieron mejoras notables en el rendimiento de sus estudiantes.
    • Esto incluye países latinoamericanos entre los que destacan los casos de Colombia y Perú.
  • El aumento de la matrícula escolar no debe realizarse con un costo a la calidad.
    • ¿Qué significa esto exactamente?  En estos países, más estudiantes de 15 años asisten ahora a la escuela, muchos de ellos muchos de ellos provenientes de ambientes socio-económicos en desventaja. Por ejemplo, en México, la cobertura de PISA era de menos del 50% de las personas de 15 años, mientas que ahora es del 65%. Contrario a los esperado, la inclusión de alumnos en desventaja no empeoró el desempeño de estos países en PISA demostrando que aumentar la matrícula escolar no significa sacrificar la calidad de la educación.

Tercero, equidad

Lograr una mayor equidad en la educación es esencial para un crecimiento incluyente.

PISA muestra que el impacto de los antecedentes sociales y económicos en el éxito educativo varía mucho de un país a otro.

  • Algunos países muestran que la equidad y la excelencia académica pueden ir de la mano.
    • Miren Australia, Canadá, Dinamarca, Estonia, Finlandia, Hong Kong (China), Japón, Corea, Macao (China), Noruega o el Reino Unido. En estos países, el desempeño promedio fue más alto que el promedio de la OCDE, mientras que la relación entre el estado socioeconómico y el rendimiento en lectura fue más débil que el promedio de la OCDE. En promedio de la OCDE, 12% de variación en el desempeño en lectura de los estudiantes puede ser justificado por el estatus socio económico de los estudiantes.   
    • Sin embargo, no podemos obviar el hecho de que ninguno de los 10 países de ALC mostró mayor equidad que el promedio de la OCDE, con la excepción de Brasil, Chile, Colombia y México, que mostraron niveles de equidad similares al promedio de la OCDE.
    • Estos resultados sugieren que los 10 países de LATAM necesitan mejorar el nivel general de rendimiento de los estudiantes sin aumentar aún más la brecha de rendimiento relacionada con el estado socioeconómico.

¿Qué se puede hacer para mejorar el rendimiento de los estudiantes?

  1. Las asignaciones de recursos a las escuelas deben ser cuidadosamente reexaminadas.
    1. En casi todos los países y economías que participaron en PISA 2018, las escuelas con desventajas socioeconómicas tendieron a sufrir más por la escasez de recursos escolares en comparación con las escuelas más aventajadas.
    1. Esta brecha socioeconómica en los recursos escolares fue especialmente grande en Perú y Uruguay.
    1. En contraste, en Costa Rica, la brecha es menor que el promedio de la OCDE.
      1. Sin embargo, la historia no siempre es sencilla. En Costa Rica, los directores tanto en las escuelas aventajadas como desaventajadas reportaron un nivel de escasez de recursos escolares más alto que el promedio de la OCDE.
      1. En este pais, es imprescindible mejorar el nivel y la calidad general de los recursos.

Cuarto, resultados sociales y emocionales.

El bienestar de los estudiantes es particularmente importante. Esto forma la base de su vida y también el desempeño en sus estudios.

  • Medir su bienestar es, por ello, fundamental.
  • Como sabemos, los estudiantes a la edad de 15 años se encuentran en una fase de transición clave de su desarrollo físico y emocional.
  • Cuando analizamos los resultados de la evaluación socioemocional, vemos que sufren muchas turbulencias emocionales.
  • Los sistemas educativos de las economías más avanzadas en todos los demás aspectos, como Singapur, Hong Kong (China) y Macao (China), son los que necesitan más mejoras.
  • Por ejemplo, los estudiantes en estos países y economías tienden a reportar más miedo al fracaso que los estudiantes en otros países / economías.
  • Puede ser tentador concluir que un rendimiento superior en la escuela conlleva una ansiedad creciente por el trabajo escolar y, por lo tanto, socava el bienestar de los estudiantes. Pero, de hecho, países tales como Estonia y Finlandia mostraron que el alto desempeño puede ser alcanzado sin mucho estrés (o un alto nivel de miedo al fracaso). Será interesante hacer un análisis más profundo sobre esto, pues sacar conclusions sencillas puede no ser el major camino.
  • Lo más preocupante es que, en un tercio de los países y economías que participaron en PISA 2018 (incluidos países de ALC como Perú, República Dominicana, México, Argentina y el Perú), más de 1 de cada 2 estudiantes reportaron que la inteligencia era algo que no podían cambiar mucho de ellos mismos.
  • En casi todos los países y economías que participaron en PISA, los estudiantes desfavorecidos tienden a creer menos en su potencial para lograr algo haciendo el esfuerzo.
  • Es una noticia triste. Es poco probable que esos estudiantes, que tienen una perspectiva tan negativa de su futuro, hagan las inversiones necesarias para tener éxito en la escuela y en la vida.
  • La ambición de crecer y progresar parece estar constantemente asociada con la motivación de los estudiantes para dominar las tareas, la autoeficacia general, el establecimiento de objetivos de aprendizaje y la percepción del valor de la escuela.
  • Además, los estudiantes que tienen la ambición de crecer tienden a reportar menos miedo al fracaso.
    • En Dinamarca, Estonia y Alemania, una gran mayoría de estudiantes informó creer en su potencial para crecer a través del esfuerzo.
  • Los encargados de formular políticas, los educadores y los padres deben proporcionar los entornos, la información, los recursos y apoyos necesarios para que los estudiantes puedan creer en su capacidad para desarrollar su inteligencia.


  • Me gustaría concluir haciendo hincapié en la importancia de las responsabilidades compartidas. Cada uno de nosotros en esta sala, como formuladores de políticas, educadores, periodistas, padres, etc., tiene un papel que desempeñar para proporcionar un entorno propicio donde nuestros jóvenes puedan crecer sin ningún sentido de límite en lo que pueden lograr.
  • En este sentido, adentrarnos en el bienestar de los estudiantes y en su relación con el entorno socio-económico junto con los resultados del rendimiento académico agregan un rostro humano a este ejercicio. 

Muchas gracias, ahora sedo la palabra al Secretario de Educación Pública de México, Esteban Moctezuma. Posteriormente estaré feliz de responder sus preguntas, tanto de los periodistas presentes como de los que nos acompañan en línea.

OECD Expert Workshop on Homelessness and Affordable Housing Policies

On the 27th of November, I delivered opening remarks at the OECD Workshop on Homelessness and Affordable Housing Policies as part of the OECD Horizontal Housing Project.

Colleagues, Ladies and gentlemen,

Thank you for joining us today.

I am glad to be opening the workshop on Homelessness and Affordable Housing Policies, which constitutes an important part of OECD Horizontal Project on Housing.  I would especially like to thank the European Commission, represented today by Olga Martinez de Briones, for their support for this work. I would also like to thank the experts and policy makers who are joining us from across OECD countries.

Homelessness is a topic that is high on the agenda of so many countries. And for a very good reason. If you walk around most major cities today, including Paris, it is impossible not to notice people sleeping on sidewalks, in metro stations, or in storefronts.

While the definition of “homelessness” differ from country to country, there are believed to be around 1.9 million homeless people in OECD countries – and this is likely an underestimate because collecting data of homeless people is extremely difficult – I come back to this later.

The situation has worsened and has become dire.

In 16 OECD countries[1], the homeless rate[2] has increased in recent years. Particularly for those living in cities, the issue has become noticeably starker in recent years. It is said that two people are dying every day on the streets of Britain, and the total number of people sleeping rough has risen by a fifth over the last year.[3]

Not to mention that children are hit the hardest. I launched the report on Vulnerable Children last week on the eve of World Children’s Day and the 30th anniversary of the United Nations Convention on the Rights of the Child. The report revealed alarming figures. Child poverty has increased in almost two-thirds of OECD countries over the past decade, with one in seven children in the OECD growing up in poverty today. The living standards of children from low-income families have also declined in many countries, particularly for those families with the smallest incomes. The report also reveals homelessness among families has risen significantly in England, Ireland, New Zealand and some US states. For children, homelessness can lead to increased anxiety, loss of contact with family and friends and poor educational outcomes.

To address this issue, we first need to know how we define homelessness and how we obtain reliable measures.

Homelessness is hard to measure.

In fact, homelessness data are hard to compare across countries. At the OECD, one of our main objectives is to provide harmonised data that allow countries to compare outcomes with their peers. But this is nearly impossible to do in the case of homelessness!


  1. First, because there is no internationally agreed upon definition of homelessness. For example, Canada, France, Ireland, Japan, and the United States use a relatively narrow definition of homelessness, which covers only people who are sleeping on the streets or living in shelters.

At the other end of the spectrum, Australia, Finland, Germany and Norway (among others) have adopted a much broader definition, which also includes people who are doubled up living with friends or family members because they don’t have a place of their own. This means that when we look at homeless estimates across countries, we are not necessarily comparing “like with like.”

  • To make it more difficult to track, homelessness isn’t always so visible. And the numbers of people without homes are often higher than we think. A growing number of people in OECD countries are sleeping in cars and parking lots, “couch surfing” or living with friends or family, because they can’t afford a place of their own.
  • And homelessness is a difficult circumstance to assess, because people experience homelessness in very different ways. In most countries, the most visible share of the homeless also tends to be the smallest: people who are “chronically homeless,” living for prolonged periods on the street, possibly suffering from poor mental health or addiction. In many countries, an even larger share is “temporarily or transitionally homeless”. These are people who experience homelessness for only a short period of time – often following a family break-up or a job loss.

But, first and foremost, we need to tackle the most basic and fundamental challenges. We need better data!

At the moment, not a single data collection method– whether we are using administrative data or point-in-time estimates such as street counts[4] can provide us with the whole picture of the problem. And this is the biggest challenge for us.

We need better data! Without it, we are lost in terms of how to address the issue. And this is why I’m pleased to know that today we will have an entire session dedicated to innovations in homelessness data and measurement this afternoon, with experts from Australia, France, Germany, the UK and the US. Without action, we risk the stability, health, well-being, safety, and security of countless individuals in our countries, failing to support them in tough times.

Then what do we know about homelessness in the OECD at this stage?

We know that the faces of homelessness are increasingly diverse. Across the OECD, we see growing numbers of homeless women, youth, families with children, seniors and migrants. Thus, there are many different experiences of homelessness, and policies must take this diversity into account. The figures in some countries for which we have data are staggering:

  • Youth represented more than a third of the total homeless populations in Norway and Australia.[5]
  • Family homelessness almost quadrupled in Ireland between 2014 and 2018.[6]
  • England (UK) recorded a ten-year high of homeless people over the age of 60 in 2018 – with the share of homeless seniors more than doubling in eight years.[7]
  • In Germany, the number of homeless refugees in 2018 represented nearly two-thirds of the total homeless population – and their share is growing.[8] 

We also know that homelessness tends to be concentrated in big cities:

  • Dublin is home to two-thirds of Ireland’s homeless population, even though it only represents about a quarter of the country’s total population.
  • In the United States, half of the homeless population is concentrated in just five states, with a quarter of the total homeless population in the state of California.

And, people become homeless for different reasons.

The OECD has identified three main drivers, and you will find these analysis in a few weeks as a Policy Brief on Homelessness in the OECD, and updates to the OECD Affordable Housing Database:

  • There are structural drivers of homelessness. This includes the lack of affordable housing. Housing is the single-largest household expenditure on average and has become less affordable across the OECD. House prices have increased three times faster than household median income over the last two decades and have risen than faster overall inflation.

This is a major challenge for individuals and households these days. We will take a deep dive into understanding how policy makers can prevent homelessness by providing more affordable housing this afternoon, with experts from Germany, Ireland, the Slovak Republic and Housing Europe.

  • There are systemic drivers. For example, a disproportionate number of youth become homeless after having left the foster care system. Some would transition out of prisons or mental health institutions without stable housing. For instance, the Abbé Pierre Foundation in France, present here today, reported earlier this year that one in four homeless people born in France was previously in foster care or known to child welfare services.[9]
  • And finally, there are individual drivers, such as a loss of job, a family breakdown, mental health or additional challenges.

What could be done about homelessness? What policy measures?

On the one hand, we need to better tailor homeless support to the diverse needs of the homeless population.

People who face financial difficulties or “hit a rough patch” may only require temporary housing support to help them get back on their feet. Meanwhile, women who are victims of domestic violence or refugees may need additional services relating to counselling, childcare or labour market support.

We also know that the most effective way to support the chronically homeless – people with more complex needs – is through “Housing First.”

Housing First approaches aim to put homeless people in permanent, immediate housing, with access to social services. 12 OECD countries report Housing First strategies at national level, and another seven at regional or local level.

However, while Housing First is becoming more widespread, most countries still regard this as a minor solution to the problem. In reality, in most places, emergency shelters remain the dominant form of homeless support. Not only is emergency accommodation expensive[10] but it is not a sustainable solution to end homelessness. Tension is rising between those favoring sweeps and those screaming for more services and more effective policy measures.

While there is no one fix-all cure, we want to highlight good initiatives to combat homelessness.

It is important to note that despite an overall increase of homelessness in the OECD, seven OECD countries[11] have experienced a decline. For example, Finland reduce their homelessness rate by 39% between 2019 and 2018! This shows that a few countries have enacted impactful policy changes in recent years.

Later today, we will hear more about the measures taken by national and local authorities in Canada, Denmark, France, Finland, Ireland and the UK.

I am sure that today’s workshop will be a great opportunity to share good practices and learn from each other.

And in the closing roundtable, welook forward to your views on how the OECD can best move forward with this work so that we can best support your efforts to end homelessness once and for all. This work will also contribute to the OECD Horizontal Project on Housing, which draws on expertise from across the Organisation to provide policy makers with guidance on the very complex issue of housing.

Thank you.

[1] Australia, Chile, the Czech Republic, Denmark, England (UK), France, Iceland, Ireland, Latvia, Luxembourg, the Netherlands, New Zealand, Portugal, Scotland (UK), the United States and Wales (UK).

[2] The homelessness rate is calculated as the number of homeless people as a share of the total population.

[3] Independent, 21 November 2019

[4] For instance, administrative data from shelters and local authorities can provide an estimate of how many people seek out support from public services over the course of a year. But they leave out people who don’t try to get public support – or don’t think they are eligible in the first place. Research has shown, for instance, that women are less likely to go to shelters, preferring instead to turn to friends and family in a first instance. On the other hand, point-in-time estimates, such as “street counts”, may better capture the share of the unsheltered homeless population. But they only provide a snapshot of the phenomenon on any given day – and fail to capture people who experience homelessness for a short time, or transition in and out of homelessness.

[5] 2019 OECD QuASH

[6] 2019 OECD QuASH

[7] Bulman, M. (2018), Number of homeless pensioners in England hits 10-year high, figures show, The Independent,


[9] Fondation Abbé Pierre (2019), L’état du mal-logement en France 2019 : Rapport annuel #24,

[10] The Guardian reported recently that England had spent over GBP 1 billion on emergency housing for the homeless – an increase of 78% in just five years (See:

[11] Austria, Canada, Finland, Israel, Norway, Poland and Sweden

10th Global Forum on Transparency and Exchange of Information for Tax Purposes

On the 25th of November the OECD hosted the 10th Global Forum on Transparency and Exchange of Information for Tax Purposes. The Secretary General and Minister Bruno Lemaire delivered opening remarks. As part of the opening session, I moderated the ministerial panel with panelists Pierre MOSCOVICI, European Commissioner for Economic and Financial Affairs, Taxation and Customs, David MASONDO, Deputy Minister of Finance, Republic of South Africa (RSA), Pierre GRAMEGNA, Minister of Finance, Luxembourg, Anne MICHEL, Journalist, Le Monde, Wolfgang SCHMIDT, State Secretary at the German Federal Ministry of Finance. See my opening remarks below.

In 2009, G20 Leaders called for the end of bank secrecy and mandated the Global Forum to deliver on that objective. As we look back on these past ten years, the world has changed and spectacular advances have been made in international tax cooperation. Bank secrecy and other barriers have been removed: the international community can now effectively tackle offshore tax evasion.

Before 2009, it was possible for wealthy individuals to hide money securely in bank secrecy jurisdictions, confident that tax authorities in their country of residence could never get access to information on those assets. While exact figures are lacking, based on the amounts recovered up to now we can safely assume that over the decades billions of USD were thus being lost for public spending. Those fraudulent practices undermined public finances, eroded tax  morale, jeopardised trust in governments and institutions, and became absolutely and simply intolerable in a context where citizens were going through the economic and social hardship of the 2008-2009 financial crisis. 

But a quantum leap has been achieved since then. Now, nearly 70 jurisdictions reviewed by the Global Forum have eliminated banking secrecy for tax purposes, some of them making profound changes to their regulatory regimes and practices since 2009. With the beginning of the automatic exchange of information on financial accounts of non-residents in 2017 and 2018, the era of bank secrecy has been taken over by that of transparency: the Secretary-General just mentioned the impressive amounts in additional revenue that since have been collected by tax authorities.

Another important change has been the multiplication of Exchange of Information instruments allowing tax information to flow between Global Forum members. The Multilateral Convention on Mutual Administrative Assistance in Tax Matters created nearly 8’000 relationships among the 130 members of the Global Forum, tripling the number of existing instruments. This network includes all G20 and OECD countries, practically all international financial centres (IFCs), and an increasing number of developing countries. There have been many positive changes as well improving the transparency of corporate vehicles. For instance, bearer shares have been eliminated or immobilised and the use of nominees is now transparent.

These successes constitute a clear example of what can be reached when the most influential economies agree on a policy objective and put their clout behind it. The process at the Global Forum is a demonstration of effective multilateralism.

How we got to this point is one of the main issues that I would like to discuss with our distinguished panellists. Was it the circumstances of the financial crisis, the leaks in the media, the need for additional revenue and public sentiment of injustice with the wealthy and international banking? Or were these changes already on track and going to happen anyway? How do we look back on those ten years and learn lessons to ensure that the same momentum can be maintained for further progress? How can we ensure the continued success of this multilateral approach?

Changing the Odds for Vulnerable Children: Building Opportunities and Resilience

On 19 November 2019, I delivered opening remarks and presented the findings of the OECD’s latest report “Changing the Odds for Vulnerable Children: Building Opportunities and Resilience” in the opening session with OECD Secretary General Angel Gurría and champions Kailash Satyarthi and HRH Princess Laurentien of the Netherlands.

Investing in children needs to be a priority for OECD countries. Childhood is a crucial moment in the development of individual. It is also a critical issue for societies and economies, as it determines the formation of human and social capital.

Across the OECD, public spending on families is low despite the potential for returns; for instance, six-times more is spent on the elderly than on families.   The potential for returns in public investment during childhood has the strongest returns of any part of the life-cycle, and returns on adult spending can be higher when there are positive spillover effects on children.

Investment in vulnerable children is most effective when it happens across the lifecycle. The factors determining the level of investment needed differs by country context. In some cases, it may mean greater investment. In other cases, it may means better investment into areas that improve value for money.

In response to rising inequalities, the OECD has called for a new growth narrative that puts people’s well-being at the centre of policy and moves beyond GDP as the sole metric of success. The OECD has developed a Framework for Policy Action on Inclusive Growth to offer countries guidance on how to design and implement policies that will give all people, firms and regions the opportunity to thrive- particularly those that are struggling or have been left behind.

This report deep dives and unpacks the first pillar (Investing in those left behind) with a focus on vulnerable children.

Current levels of income inequality are framing children’s life chances.

In OECD countries, income inequality has been growing rapidly; the average disposable income of the top 10% is now around nine and half times that of the poorest 10%, up from seven times 25 years ago. Globally, the share of gross national income held by the top 10% continues to grow and is particularly marked in emerging economies of India and Russia.  The more unequal a country is, the lower the rate of social mobility. In low-inequality countries such as the Nordic countries it would take 100 years or four for children born into low-income families to reach the average income and in high-inequality countries like such as some of the emerging economies- Brazil, Columbia and South Africa, it would take nine generations, or even longer. 

The share of middle-income households with children is in decline. It has fallen from 72% to 68% for couples with children, and from 55% to 44% among single parents.  Between the mid-1980s and mid-2010s, single-parent households increasingly dropped out of the middle-income group and, by the end of the period, a majority were on lower incomes.

Middle-income spending is rising faster than income and the middle class are spend more than half of their budget on core items such as housing, food, clothing, health and education. Housing is the largest spending item which increased from ¼ in 1995 to 1/3 in 2015.  In some countries, the rising costs of higher education threaten the ability of the middle-income class to send their children to university, as they have increased faster than inflation and median incomes.

Many middle-income households are financially vulnerable or struggle to make ends meet. Households with children are the most economically vulnerable meaning they lack the liquid financial assets needed to maintain a living standard at the poverty level for at least three months .

Children are vulnerable for different reasons. This report considers child vulnerability as the outcome of the interaction of a range of individual and environmental factors that compound dynamically over time. Types and degrees of child vulnerability vary as these factors change and evolve. For example, age shapes children’s needs while also exposing them to potential new risks. The independence of older adolescences makes them more susceptible to opportunities and risks in the community, making the presence of supportive adults, school quality and local economic opportunities important for well-being.

Individual factors include disability, mental health difficulties, immigrant background, experiencing maltreatment or being in out-of-home care. For instance, children with disabilities are more likely to live in low socio-economic households, to experience maltreatment and to be bullied. They are overrepresented in institutional care.

Environmental factors operate at the family and community level. Family factors include material deprivation, parents’ health and health behaviours, parents’ education level, family stress and intimate partner violence. Community factors are associated with school and neighbourhood environments.

Family factors shaping child vulnerability include material deprivation. OECD broadly defines material deprivation as the inability of a household to afford consumption goods and activities that are typical in a society at a given time.

For example, nutrition-  One in ten children in European OECD countries do not have access to a healthy diet as measured as access to fresh fruit or vegetables and/or one meal including meat, chicken, fish or vegetarian equivalent once a day. In low income households, the food budget is most likely to be cut.  Access to cooking and food storage facilities and locally available food options also determine children’s diet. Schools and after-school clubs can play an important role in supplementing the diet of vulnerable children. Poor nutrition negatively affects child development and health and interferes with children’s ability to preform well at school.

One third of children in European OECD countries experience deprivation in leisure activities as measured as not being able to participate in a regular leisure activity and/or go on a holiday away from the home at least one week a year.  The rate is two-times higher among income-poor children.

Participation in regular leisure activities help children develop social skills, friendships, and positive subjective well-being, and is associated with improved educational outcomes. Leisure activities are especially important for vulnerable children, as they provide natural opportunities to interact with supportive adults and mentors, as well as time away from stressful home environments.

Moreover, one in six children in European OECD countries experience severe deprivation measured as being deprived across four of seven dimensions. The risk of severe deprivation is closely related to poverty. On average, 36% of children living in poverty experience severe deprivation.

Intimate partner violence in the home is increasingly recognised as serious problem in OECD countries.

At the EU level, 22% of women report physical and/or sexual assault by their current or previous partner, and 43% disclosed psychological abuse. 73% of women reporting IPV are caregivers of children who also witness the abuse.  The life-time exposure of children to intimate partner violence ranges from 14% in Sweden to 28% in the United States.

Intimate partner violence has serious consequences for child well-being. Exposure during pregnancy is associated with low birth weight and pre-term delivery. In early childhood, it can have long-term consequences on social and emotional development. Children affected by intimate partner violence benefit from the risk of exposure to intimate partner violence being reduced and/or eliminated, but also from interventions to strengthen parent-child relationships.

In developing countries, where a child lives and household wealth determines access to basic rights and services. Far too many children do not have their birth registered. Birth registration is a prerequisite for accessing government services and social protection, and provides protection from exploitation and access to juvenile justice.

Birth registration is 30% lower in rural areas.

Children from the wealthiest families are 1.5 times more likely to have their birth registered than children in the poorest households.

Furthermore, children in the poorest households are much less likely to have access to a favourable learning environment at home or through an Early Childhood Development Programme. In least-developed countries, an estimated 29% of children in the wealthiest quintile were enrolled in a programme compared to only 7% of children in the poorest quintile.

Child labour remains a critical children’s rights issue. Child labour has declined but progress has slowed. In 2016, about one in ten children aged 5-17 years engaged in some form of child labour, and nearly half in hazardous forms of work. About nine in ten children in child labour live in Africa or in Asia and the Pacific region. In Africa, one in five children are in child labour. 

The OECD is recommending countries to approach improving the well-being of children through cross-cutting child well-being strategies with a particular focus on vulnerable children, and deliver policies that develop these children’s resilience.

Vulnerable children need consistent, coherent and coordinated support throughout childhood and this requires a whole-of-government approach to child policy.

The OECD has put forward six areas of policy action around which child well-being strategies could be organised.

These policies reduce risks and increase protective factors, thereby building children’s resilience. Protective factors mitigate risk and reduce negative outcomes. They allow children to benefit form positive experiences , form key capabilities, and access resources in favour of good outcomes. Protective factors are present in the family and the community; some are embedded in relationships children have with adults and others through local resources such as effective schools and neighbourhoods and strong child protection systems.

For children who experience high levels of adversity, enhancing the quality of children’s environments and making available resources to nurture and sustain well-being are the most important interventions.

Improve children’s educational outcomes by increasing vulnerable children’s participation in ECEC. Participation in early childhood education and care can be an important protective factor in the lives of vulnerable children. Vulnerable children access ECEC at a much lower rate, in some countries up to half.

PISA 2015 show that 15 year-old students who attended ECEC for two years or more score a significant 26 score-points higher in science test the PISA test assessing sciences performance than their counterparts who attended ECEC for less than two years. This is roughly worth half an academic year of schooling.  The magnitude of benefits depends on the quality of ECEC. High quality ECEC can have a stronger positive effect on children whose mothers have lower-level of education.   Evidence suggests that ECEC can help parents engage more frequently in cognitively stimulating and less passive activities, helping close the gap between disadvantaged children and children from non-disadvantaged families. 

Many countries gave already have integrated social and emotional skills development into their national and sub-national curricula. For example, Norway has introduced building life-skills and learning about mental health as a cross-curricular theme. Ireland has introduced a framework for use in early years and primary school setting to promote well-being and a sense of identity and belonging.

Policy Action: Strengthen child protection.

One area for greater investment is in improving outcomes for children in out-of-home care as the outcomes for these vulnerable children is much lower than the general population, across education, health, adult employment and futures earning.

Enhancing the well-being of children placed in out-of-home care requires greater investment in resources that build protective factors.

We know that children do better when they come into care at a younger age, have minimal care and school placement disruptions, when they are placed in kinship or foster care, when they are supported to have positive contact with their birth family, and when the receive support on leaving the system. We need to have policies in place that support better outcomes. For example, policies are needed to support positive contact with children and their birth family.  Child protections services can influence the quality of contact by supporting parents and children by supporting parents and children during and between contact visits to make full use of this time .

Countries should have in place policies to support the transition of young people ageing out of the care system. This entails a statutory entitlement for support until reaching a particular age and/or completing education. For example, New Zealand offers young people the option of remaining or returning to live with a carer until 21 years of age and to access transition support and advice until 25 years of age.

We need to better understand the policy determinants of child resilience. We need to understand which protective factors contribute more to positive well-being.

OECD can take on an active role in designing a child well-being framework and ensure the cross-national comparability of the definitions used to categorise family situation and to compare children outcomes.  New Zealand is one example of a country that is starting to implement a child well-being strategy close to this set of policy actions outlined in the report Changing the Odds for Vulnerable Children. 

Conference de Paris: Maintaining a Multilateral Dialogue

On 14 November 2019, following the opening of the second day of the Conference de Paris by Mayor of Paris Anne Hidalgo, I participated in a panel entitled “Maintaing a Multilateral Dialogue” with José Luis Manzano, President, Integra Capital and Jean-Hervé Lorenz, President, Le Cercle des Économistes. Find my panel remarks below:

Ladies and Gentlemen,

  • It is again a pleasure to host this important Forum at the OECD.
  • It is revealing about the global context we are in that this session is devoted to maintaining multilateral dialogue. We need more ambition!
  • The reality is that the political, economic and environmental systems that govern our lives are under enormous strain, and just when we need it most, the global order is fracturing.
  • Countries are struggling to form governments, and populist and protectionist politicians are increasingly marking the political landscape.

‘Deglobalisation’ is bringing huge costs

  • Between 1991 and 2017 global trade quadrupled.
  • Growth of global exports in business services and financial services grew by five times, and manufacturing tripled.
  • Due to global value chains, the OECD has estimated that 40% of jobs in the OECD area are dependent on foreign trade.
  • Almost 5 million jobs in the US are dependent on trade with Mexico and 40% of the content of US imports from Mexico consists of US value added.
  • But what we are seeing now is an attempted deglobalising and protectionist trends.
  • Trade tensions and protectionist policies are bringing worrying consequences:
    • According to OECD Economic Outlook global growth will slow to 2.9% in 2019 and 3% in 2020: weakest annual growth rates since the crisis.
    • Trade growth has collapsed below 1%.
    • New trade restrictions in G20 countries in 2nd half of 2018 affected almost 500 billion US dollars of imported products.
    • In G20 countries aggregate investment growth has dropped from 5% at the start of 2018 to only 1% in the first half of 2019.
    • Global industrial production has fallen below 2%. In Germany it has dropped almost 10 percentage points to around -5% in 2019!
    • The OECD estimates that the impact of US-China trade restrictions could lower global GDP by 0.7 percentage point per year in the first two years of the shock and global trade growth by close to 1½ per cent per year, with the effects felt all over the world.

Economic trends combined with inequalities are eroding trust and fuelling anger

  • Over the last 30 years, despite deepening global integration, inequality has risen in many countries, meaning the benefits of global growth have not been shared fairly.
    • The richest 10% used to earn seven times more than the poorest 10% in the OECD, this ratio is now around nine and a half times.
    • In the United States, for example, the share of the top 1% has almost doubled from about 11% to 20% and almost half of all income growth accrued to this group.
    • Median incomes have not followed suit. Over the last 30 years median incomes in the OECD increased a third less than the average income of the richest 10%.
    • Inequality of wealth is even more pronounced: the top 10% holds half of total wealth while the bottom 40% holds only 3%.
    • Economic inequalities translate into social divides, with compounded effects: by the age of 15 disadvantaged pupils in the OECD have fallen on average two-and-a-half years behind their more affluent peers.
  • All these factors have driven down trust: In the OECD only 43% of citizens trust their government.
  • There is also not enough progress tackling gender divides: pay gap is still 14% in OECD, and women spend double the amount of time in unpaid work.
  • The ILO estimates that globally 606 million women, or 41% of those currently inactive, are outside the labour market because of their unpaid care responsibilities.

Our growth model has left people behind but also created an environmental emergency

  • Emissions have started to rise again, reaching unprecedented levels in 2018. We are not on track to reduce warming to below 1.5 degrees, our kids are striking, the Amazon is burning, our oceans are being suffocated by plastics.
  • By the middle of the century, there will be more plastic than fish in the oceans, by weight.
  • Scientists warn that a million plant and animal species face extinction while the health and security of billions of people are at risk.
  • By 2025 half of the world’s population will be living in water-stressed areas.
  • The OECD has estimated that outdoor air pollution could cause 6 to 9 million premature deaths a year by 2060.

There is also anxiety around megatrends like digitalisation, which could deepen inequalities

  • The OECD estimates that around 14% of jobs are at high risk of automation, and another 32% will be changed significantly, with the low-skilled and low-paid most at risk.
  • Yet low skills workers are, on average, 40 percentage points less likely than high-skilled adults to participate in training.
  • Middle-skill occupations are disappearing fast.  On average in 21 OECD countries with data, the years between 1990 and 2010 saw middle-skill occupations losing 8% employment share.
  • We are also seeing emergence of non- standard jobs, particularly in the platform economy. These workers are up to 50% less likely to get income support when out of work.
  • It’s not just about social costs, there are also anxieties around concentration at the top: only 250 firms globally generate 70% of R&D and patents, and 44% of trademarks. All too often, they are also not paying their fair share of tax.

We need to reshape the foundations of multilateralism and put people at the centre

  • The OECD is advocating people-centred growth (we have the Policy Framework for Action on IG, Gender Strategy, PISA, Well-Being Framework, SDG Action Plan & Governance Hub) and we have NAEC.
  • We have the Going Digital Project & AI Principles.
  • We are also creating synergies between climate action and well-being (quality jobs, health, skills)
  • We are tackling tax evasion and avoidance, with BEPS and Automatic Exchange of Information (95 billion euros in additional revenues).

Paris Peace Forum: A Fusion of Solutions: Towards Common Principles on Artificial Intelligence?

On 13 November 2019, I participated in a panel entitled “A Fusion of Solutions: Towards Common Principles on Artificial Intelligence” with Kersti Kaljulaid, President of the Republic of Estonia, Nicholas Théry, President of Crédit Mutuel, Yeong Zee Kin, Assistant Chief Executive, Infocomm Media Development Authority of Singapore, Sasha Rubel, Programme Specialist, Knowledge Societies Division, UNESCO. See my panel speaking points below:


  • AI is transformative in healthcare, learning, public services, safety, transport, sustainability…
  • For example, MIT professor Regina Barzilay, is developing AI to analyse mammograms to detect cancer earlier, after a screen missed her illness.
  • By identifying patterns, AI can also contribute to real-time detection of financial markets fraud.
  • AI optimisation can drive more efficient energy use and transportation planning & reduce menial tasks.
  • AI is growing: AI start-ups attract around 12% of global private equity investments & AI patents grew 10x between 1990 & 2016 (3x for all patents).[i]


  • AI systems work based on patterns detected in datasets, so they risk reinforcing existing biases in society.
  • Gender biases: A study found, for example, that men were almost six times more likely than women to be shown ads for high-paying executive jobs.
  • Racial biases: Research has shown that AI systems sold by tech giants have error rates of max1% for lighter-skinned men and 35% for darker-skinned women.[ii]


  • AI can infringe on human rights and privacy: risk of data manipulation & sharing; identification and tracking; speech and facial recognition. Evidence of AI tech being used by governments to spy on civil society and activists.
  • AI can infringe on democracy: algorithms and ‘bots’ help share fake news, violent images, harvest & sell data (cambridge analytica) and drive echo chambers, esp on social media.


  • We need to avoid multiplication of approaches: currently national AI strategies, AI ethical codes by businesses, AI standards from technical community.
  • OECD has counted 26 sets of AI guidelines developed by stakeholders.
  • No single country or category of actors has all the answers to these challenges. We need a global multi-stakeholder response to a global issue.
  • We need to speak the same language & develop sound evidence-based approaches to common AI goals.


  • This year 36 OECD Members and 7 partner countries have adopted the OECD Recommendation on AI, the first intergovernmental set of AI Principles.
  • They served as the basis for the G20 AI Principles.
  • The OECD AI Principles represent a global reference for responsible stewardship of trustworthy AI.
  • They identify shared values-based principles: sustainable growth & well-being; human-centred values and fairness; transparency & accountability.
  • They recommend specific policies for international co-operation, to foster alignment around technical standards, metrics and interoperability.


  • OECD AI Policy Observatory is a multi-stakeholder, multi-disciplinary hub to provide guidance on implementing the AI Principles, policy analysis and resources, and indicators and measurement work.
  • It will include a digital library of countries’ and stakeholders’ AI initiatives to help mutual learning: so far 270+ policy initiatives from 40+ countries.
  • Given the global nature of AI governance and thanks to its broad reach and inclusive approach, the AI Policy Observatory will be integral to the international debate on AI policy solutions.


  • OECD called upon by France & Canada through the G7 to support the Global Partnership on AI.
  • GPAI will work with OECD to guide the responsible adoption of AI grounded in human rights.
  • There is great scope for convergence between AI initiatives globally: we all recognise the need for fairness, transparency, and accountability.
  • This consensus is crucial, but we cannot rest there. We need to raise our ambitions for concrete actions and implementation mechanisms, like the OECD AI Policy Observatory that will facilitate AI global governance in the years to come.

[i] OECD (2019), Measuring the Digital Transformation: A Roadmap for the Future, OECD Publishing, Paris.


Pitch of the NiñaSTEM Pueden Project

At this year’s Paris Peace Forum, NiñaSTEM Pueden was exhibited as a selected “Inclusive Economy” initiative. On 12 November 2019, I presented the pitch of this simple yet effective initiative to combat negative gender-based stereotypes that hold women back from pursuing STEM in Mexico. We brought this project to the Paris Peace Forum to forge important connections both for our project in Mexico and with other countries who may wish to implement a similar initiative. The OECD Secretary-General attended this pitch.

Ladies and Gentlemen,

My name is Gabriela Ramos; I am the OECD Chief of Staff and the Sherpa to the G7 and G20, and I also oversee the work of the OECD on gender. I am delighted to present to you today a project that is very close to my heart, because it is helping transform the lives and opportunities of girls in my country, Mexico.

But let me begin, with a little bit of context.

At the OECD, we have been working hard since the 80s, but with renewed efforts since 2011, to promote gender equality.

As our mandate calls for “Better Policies for Better lives”, we have focused on policies and on interventions that matter. We have encouraged countries to legislate to ensure gender equality, and to review legal framework; we have promoted affirmative action for management and leadership positions; and we have adopted family-friendly policies to share unpaid care work more equally.

And we have seen progress, with the G20 adopting a gender target, with many countries repealing damaging laws, and with many countries, including in the OECD, adopting or improving dual parental leave policies.

However, this progress is too slow, and as UNWomen has said, it will take us 200 years to reach parity if we continue with this pace.

This is a human right story, it is about representation and access, but it is also an economic one, and if we think about the gender target in the G20 we understand why.

Increasing labour force participation by 25% will bring 100 million women into the labour force and increase substantially GDP in a context of slow global growth. According to the OECD’s Social Institutions and Gender Index (SIGI) the current level of discrimination reduces global income by 7.5%, a loss of USD 6 trillion.

So, if we know the tools, and we know the policies. If we are convinced this is the right thing to do, why is it so difficult? Well, it has to do with our cultural norms and understanding, sometimes with religious believes or moral frameworks, but sometimes by plain stereotyping that put women in certain role and men in another one.

It starts in the family, where we promote the idea that girls should not dare, that they are there to be protected, or by the media, including social networks, that objectivize women. There is even acceptance by some women that violence is justified!

What is the outcome of this stereotpying? Parents not thinking their girls can reach the sky, girls thinking that they are not good enough and with lower confidence (10% lower according to PISA).

All this translates into under-representation in STEM. Women make up more than half of university graduates, but not even a third of STEM graduates, and only a fifth of computer science graduates.

So at the OECD, inspired by Chancellor Merkel, we decided to launch a program to break this stereotyping, that is really not expensive, but that can have great dividends.

The story of the project is straightforward. In 2016, when Chancellor Merkel chaired the G7, she promoted the idea that we needed more role models.

She invited a group of prominent women from different walks of life. The idea was that visibility could send a strong message. For me it was even clearer at a personal level, I remember when the OECD’s gender expert told me her daughter asked her whether a man could be a chancellor!

So we borrowed this idea for Mexico, where only 9% of Mexican women enrol in STEM majors.  And launched something very simple.

We looked for successful women in Mexico in the areas of STEM. And we found amazing people.

We found Julieta Fierro, an astronomer and leading research scientist; Dorothy Ruiz-Martinez, a NASA engineer; Deborah Berebichez, the first Mexican women to earn a PhD in physics from Stanford.

We gathered 50 STEM “mentors” and the Public TV in Mexico produced clips to encourage girls to aspire to STEM. The previous Minister of Education endorsed it, and the new administration too. It is the Ministry of education that opens the doors to the schools, and gathers the girls.

The mentors speak with girls that are about to choose their studies, and make them consider this option. The mentors do not ask them if they want to be engineers, they ask them if they want to build robots. Everybody wants to build robots! Everybody wants to go to the space! Everybody wants to clean the ocean with new technology, or save lives.

The results are incredible. NiñaSTEM has reached more than 2,500 female and 1,000 male students between 7th-9th grade.

And now we are replicating this simple format in other countries, and we want you to vote for this program, to be able to promote it beyond. Certainly, this will not resolve all the problems of stereotyping. But it brings confidence to the girls, and to the parents and teachers, it shows how the biases and cultural barriers that impede progress.

Of course all this has to come with additional policy reforms: eliminating gender stereotyping from text books, and take action particularly in social networks to avoid girls to be objectivized, and appreciated only for their looks for example.

But most girls will only become what they can see is possible, and this is the mission of NinaSTEM Pueden!

Paris Peace Forum Panel: The Challenge of Alignment Towards a Common Framework for SDG-Compatible Finance

On 12 November 2019, I participated in a panel entitled “The Challenge of Alignment Towards a Common Framework for SDG-Compatible Finance” with M. Abdoulaye Mar Dieye, Special advisor, UNDP; M. Mathias Vicherat, Secretary General, Danone ; Mme Aba Esther Eshun, young expert, AU-EU Youth Cooperation Hub and a respresentative from the Secretary of State to the Minister of Europe and Foreign Affairs. See my remarks below:

Ladies and Gentlemen,

Let me get straight to the point: we run the risk of defaulting on our promise to finance the 2030 Agenda. The annual financing gap for the SDGs stands at 2.5 trillion dollars in developing countries alone.

And yet, our Global Outlook on Financing for Sustainable Development concluded that total external support to financing the SDGs in developing countries had actually dropped over the past year, primarily due to a contraction of FDI, which has declined by 20% in the first half of 2019.

At the same time our countries have been struggling to maintain the level of official development assistance (ODA) to a collective $150 billion. We would need 20 times more to fill the gap.

We need to “shift the trillions”, as the French G7 Presidency put it, with the Development Ministers’ Declaration calling for a more comprehensive approach to SDG-compatible finance, based on the work of the OECD and the UN, and other stakeholders. For this, we need a strategy oriented around three key pillars: mobilisation, alignment, and impact.


The financing is there, if we could only mobilise it. That begins with increasing domestic resource mobilization, which remains the primary source of financing for the SDGs.

We know that increasing the tax to GDP ratios across developing countries by just 1% would provide an extra $250 billion in revenues in developing countries.[1] 

Our joint program with UNDP, Tax Inspectors without Borders (TIWB), creates $100 of tax return to developing countries for each dollar spent, and has already collected around 500 million dollars of additional tax revenue since it was launched 2015.

On the global scale, OECD work on Automatic Exchange of Information has helped raise close to EUR 100 billion of additional tax revenues. Base erosion and profit shifting, which we are tackling through the OECD/G20 Inclusive Framework on BEPS, disproportionately affects developing countries, and costs up to 240 billion USD per year in lost tax revenue.

However, in addition to taxation, we must also increase ODA volumes from 0.3% of GNI in DAC countries to the UN approved target of 0.7%.

We also need to mobilise new forms of financing. The 2.5 trillion USD gap actually represents less than 1% of available global financial assets. But, as the G7 development ministers recognized, we have to find new solutions, like blended finance.

Some progress has been made. DAC members mobilized over $150 billion of private finance for development between 2012 and 2017 through the use of blended finance and other instruments. Impact investment is another example, the estimated total global impact investments stand at approximately USD 502 billion, but it’s still not enough.


We are lagging behind in mobilizing, but we are lagging three times behind in alignment.  

The size of the challenge would require to re-think all the budget processes and allocations to align them with the SDGs, but we are not doing it nearly enough. Less than half of OECD countries specifically include SDG reporting in their budgets.

New Zealand is interesting because they moved to Well Being budgeting. This is based on the idea that gauging the long-term impact of policies on the quality of people’s lives should be the focus of the national budget, rather than focusing on short-term output measures, and not taking account of the negative consequences of policies.

And more than anything, we need to stop dedicating resources to the ‘wrongs’. Spending more in fossil fuels (almost 1 trillion) instead of financing renewables, for example.

The OECD is launching a new report on the alignment of development co-operation with the Paris Agreement, to end the practice of using concessional finance for fossil fuels and instead to finance low-carbon alternatives.

Alignment is about incentives, financial markets cannot continue rewarding the same pattern of consumption and production that has led to such social and environmental outcomes.

Mark Carney talked about the “tragedy of horizons”, imposing a cost on future generations that we have no incentive to fix, and pointed out that meaningful action would leave most of our assets “stranded”.

The key will be to reward and encourage good investment decisions and for that we need to change the metrics.

Instead of calling for more FDI without discriminating, we should call for quality FDI.

Just the other day, I launched the OECD FDI Qualities Policy Toolkit, which will advise policymaking to improve alignment of FDI with the SDGs.

The indicators focus on five clusters linked to the SDGs: productivity and innovation; employment and job quality; skills; gender equality, and carbon footprint.

These indicators will be a practical tool to shape the investment policy framework conditions to deliver on the SDGs.  Because this is ultimately about impact, and impact requires measurement.


We need to change the metrics to overhaul existing financial and development systems. This means incorporating   impact measurement – including both positive and negative outcomes.

I met with Sir Ronald Cohen at the UNGA in September . We discussed that the criteria for business and measurement should not only be risk and return, but include impact. The same for the traditional cost benefit analysis by our governments.

But this requires a common framework for impact measurement.

This is why the OECD is working with partners like UNDP and the IMP Structured Network to develop a generally accepted Conceptual Framework that will contribute to impact performance measurement and integration of impact into investment decision-making by using the Financial Accounting conceptual framework (IFRS) as a basis.  

The framework will include a set of principles, with accompanying implementation guidance, regarding how information should be collected and disclosed by organisations to inform a range of decisions, including to decrease negative impact and increase positive impact on people and planet.

This will be critical in delivering on the SDG agenda, and as Betrand Badré would say enabling “finance to save the world.”

Thank you.

[1] Overall, the ratio for the 53 developing countries in the OECD Revenue Statistics is 19.1%, compared to 34% for the OECD.