COVID-19 in Colombia: Impact and Policy Responses


In this note, we review Colombia's handling of the COVID-19 crisis. The first line of policy response slowed down the pace of contagion and avoided excess deaths, providing additional time to strengthen the health system and increase ICU capacity. However, the challenges that remain are significant. Testing and contact tracing capacity is inadequate, and ICU occupancy rate is in red alert. There is lockdown fatigue, driving people back to the streets despite the virus expansion. Timely adopted economic measures have been insufficient to avoid a major recession and a significant reversal in poverty and inequality. We provide some policy recommendations for the next stages of the pandemic.


There is some consensus now that the impact of COVID-19 worldwide will be greater than initially expected. According to the IMF, the pandemic, and the policies that were implemented to control it, left larger-than-anticipated damage to the supply potential, and it will continue to have a negative impact throughout 2020 due to persistent social distancing and lower productivity from surviving businesses that need to implement workplace safety and hygiene procedures. Moreover, countries that have not been able to break the back of infection rates will have to implement longer lockdowns, which will have a greater toll on economic activity. In its June revision, the IMF’s World Economic Outlook expects that the world economy will experience a sharper contraction in 2020 (4.9 percent vs 3.0 percent projected last April), and a milder recovery in 2021 (5.4 percent vs 5.8 percent last April).

Unsurprisingly, because of the evolution of the virus, Latin America and the Caribbean (LATAM) is expected to be one of the regions worst affected by COVID-19. The IMF projects a 9.4 percent contraction for the region in 2020, in contrast with a 3 percent contraction for emerging markets and developing economies and an 8.0 percent contraction for advanced economies (Figure 1). Likewise, the recovery in 2021 will be weaker.

Figure 1. GDP annual forecast (percent) – IMF WEO updated JUNE 2020

Country Groups

LATAM (Selected Countries)

Source: IMF WEO update June 2020

Among the major economies in LATAM, the “top” performers will be Chile and Colombia, which are expected to decline 7.5 percent and 7.8 percent, respectively. On the other end of the spectrum, the IMF has Mexico and Peru contracting by more than 10 percent. In this policy note, we focus on Colombia’s policy response and performance during the COVID-19 pandemic.

COVID-19 has exhibited a slower contagion pace in Colombia …

Colombia has 190,700 confirmed COVID-19 cases, 85,836 recovered cases, and 6,516 deaths (data for July 18, 2020).[1] In line with global trends, 65 percent of active cases are people aged 18-50 years. The total case fatality rate (CFR) is 3.5 percent, with a wide heterogeneity between age groups.[2] For those older than 60, the CFR is greater than 10 percent, climbing up to 33 percent for people 80 or above.

Relative to other countries, COVID-19 in Colombia has exhibited a slower but constant contagion pace. Almost 100 days after the onset of the pandemic (which corresponds to the day on which a particular country reached 0.1 daily deaths per 1 million inhabitants), Colombia now has more than 100 daily cases per million inhabitants (7-day moving average). It took about 20 days for Spain, Italy, and the US to reach this number of daily cases, and less than 60 days in Brazil, Chile, and Peru. Mexico is below the 100 per million daily cases, while Uruguay is the region’s most successful case for reasons that are not entirely clear (Figure 2).

Figure 2. Daily cases per million inhabitants (7-day moving average)

Note: Dash lines limit the area in the graph where, for each day, you find 25 percent to 75 percent of the countries in the dataset
Source: Own calculations with data downloaded from Max Roser, Hannah Ritchie, Esteban Ortiz-Ospina and Joe Hasell (2020),, 07-19-2020

…which has translated to fewer deaths

Brazil, Mexico, Peru, and Chile observed about five daily deaths per million around 60 days after the beginning of the pandemic. Colombia is below that fatality rate, reflecting the slower pace of the spread of the infection (Figure 3). Latin American countries have had lower daily deaths compared to Italy and Spain (which had peaks between 10 and 20 per million within 30 days after the beginning of the pandemic but managed to reduce deaths to 1 per million 120 days after the beginning of the pandemic). Yet, 100 days after the onset, Colombia’s fatalities (per million) are still growing.

Figure 3. Daily deaths per million inhabitants (7-day moving average)

* Following methodological changes, the total number of deceased in Spain changed in -1,918 in May 25th and 1,179 in June 19th
**Dash lines limit the area in the graph where, for each day, you find 25 percent to 75 percent of the countries in the dataset
Source: Own calculations with data downloaded from Max Roser, Hannah Ritchie, Esteban Ortiz-Ospina and Joe Hasell (2020),, 07-19-2020

Another popular measure to compare the devasting impact of COVID-19 between different countries is excess mortality.[3] The Financial Times has calculated this measure for several countries. Estimates for Brazil, Chile, and Peru suggest 54,700; 9,400; and 40,100 excess deaths, respectively. In Italy, Spain, and the US, excess deaths are around 48,600; 48,500; and 149,200, respectively. In contrast to these numbers, Colombia’s National Statistic Department (DANE for its acronym in Spanish) has provided evidence indicating the absence of excess deaths due to natural causes during the first half of 2020.[4]

In both daily deaths and excess mortality, Colombia has fared better during the first 100 days since the relative outbreak of the pandemic than similar countries of LATAM. However, there are worrying signs that suggest that this could change.

First line of defense: Testing, lockdown, contact tracing, and ICUs

As with most countries, Colombia responded to COVID-19 by increasing testing capacity, mandating strict lockdowns, introducing contact tracing, and augmenting its ICU capacity.

Colombia’s rate of daily PCR testing has increased consistently since the onset of the pandemic. The number of labs capable of processing PCR tests (polymerase chain reaction tests, used to determine whether someone is currently infected with COVID-19) increased from one to 93, which allowed daily tests to grow exponentially. Testing levels are higher than in Mexico and Peru, per million inhabitants, but lower than Chile’s. Chile and Colombia are the two countries in LATAM were daily PCR testing is in line with what per-capita income would project (see Figure 4).

Figure 4. Testing capacity

Note on graph on the right: For the vertical axis, we take the highest observed value for each country’s daily testing – 7-day moving average series. The graph is in logarithmic scale in both axes, we exclude the tails of the distribution (10 percent). The latest available GDP per capita data for each country was taken from the World Bank’s WDI dataset.
Source: Own calculations with data downloaded from Max Roser, Hannah Ritchie, Esteban Ortiz-Ospina and Joe Hasell (2020),, 07-19-2020

The Colombian government was also quick to respond by implementing a strict lockdown that began the last week of March. According to the Containment and Health Index built by the Oxford COVID-19 Government Response Tracker team, Colombia’s lockdown was imposed earlier, relative to the severity of the pandemic, than in other countries, and it continues to be one of the strictest lockdowns in the world (Figure 5).

Figure 5. Containment and Health Index

Containment and Health Index: school closing, stay at home requirements, public information campaign, and others.
Source: own calculations with data from Oxford COVID-19 Government Response Tracker, 07-14-2020

The initial effectiveness and timing of Colombia’s lockdown is confirmed by Google’s mobility data (Figure 6). At its peak, Colombians spent 30 percent more time at home, relative to the pre-COVID period. The magnitude of the retrenchment at home is among the highest in LATAM and advanced economies (except for Peru, where people spent 40 percent more time at home). In LATAM, mandates increased time at home some weeks before the pandemic had caused 0.1 deaths per million inhabitants, while in the US, Italy, and Spain, people quarantined weeks after this threshold was reached. Thus, lockdowns in Colombia, and LATAM more broadly, began before the pandemic gained speed.

Figure 6. Frequency time spent at households

Source: Own calculations with data taken from Google Community Mobility Report, data downloaded 07-19-2020

Contact tracing allows governments to preventively isolate people who have had contact with positive cases. Tracing has been a key policy component of all countries that have successfully managed to control the expansion of the virus. Colombia, Chile, and Peru, according to the Oxford COVID-19 Government Response Tracker team, all implemented a comprehensive tracing policy. Mexico, in contrast, has a much more limited tracing policy while Brazil has no specific federal tracing program.

Regarding an increase in ICU capacity of the health system, Colombia ICUs rose to 7,950 by mid-July from 5,400 before the beginning of the pandemic. This has allowed the occupancy rate, at the national level, to remain below 75 percent (67 percent on July 16).[5]

But there are some worrying signals

Despite initial efforts by the Colombian government, recent trends in the expansion of COVID-19 suggest that efforts to control the virus are entering a critical phase.

The test positivity rate continues to increase. Despite efforts to increase testing, COVID-19 has expanded at a faster rate than testing. The positivity rate has been consistently above 10 percent during the last 60 days and has a positive trend (Figure 7). Moreover, some recent reports indicate that it is hovering around 25 percent.[6] Although there is no clear answer to how much testing is necessary, the World Health Organization suggests that countries do extensive testing in order to keep positivity rates at or below 10 percent.[7] Without enough testing, it is difficult for a country to identify and quickly isolate positive cases.

Figure 7. Positivity rate (7-day moving average)

*Note: These are numbers for PCR testing. Some countries used other type of tests to confirm positive Covid-19 cases. It is possible to have more positive cases than tests (i.e. Peru)
Source: Own calculations with data downloaded from Max Roser, Hannah Ritchie, Esteban Ortiz-Ospina and Joe Hasell (2020),, 07-19-2020

Initial evidence suggests that the effectiveness of contact tracing has fallen consistently since the onset of the pandemic. Two proxy indicators of the effectiveness of contact tracing are the percentage of cases that have been identified through this policy, and the number of days that lapse between initial symptoms and diagnosis (Figure 8). By July 18, only 10 percent of the total cases in Colombia were identified by contact tracing. Using data for Colombia, Juliana Villanueva, Silvana Zapata, Zulma Cucunubá, and Andrés Vecino provide preliminary estimates indicating that an increase of contact tracing up to 50 percent of identified cases could reduce mortality rates between 7 percent and 19 percent.[8]

Figure 8. Contact tracing indicators

*The graph on the right shows the evolution of the median of days between symptoms and diagnosis (7-day moving average)
Source: Own calculations with data from INS, 07-18-2020

Additionally, the waiting period between symptoms and diagnosis reached a minimum level by mid-May but, since then, has been increasing. With the data from the July 18 report, the median length of this waiting period was 11 days. A longer waiting time makes it harder to trace contacts and enforce a well-timed isolation. More lag time means more contacts for those potentially infected.

The recent pace of COVID-19’s expansion has put the health system in red alert despite the increase in ICUs capacity. On July 16, 13 departamentos out of 33 had ICUs occupancy rates above 65 percent. [9] The capital district (Bogotá) and Antioquia, the two most populous, had occupancy rates around 80 percent, which led local authorities to reinstate strict lockdowns. Colombia’s ICU capacity is expected to continue to increase in August and September. However, the expansion will be constrained not by the number of ventilators that can be purchased, but by the number of specialized doctors that can support these facilities. Colombia is now at a moment where the health system is understaffed in terms of professionals.

Lockdowns seem to have declining effectiveness. Google mobility data provides suggestive evidence that, independently of a government’s lockdown policy, compliance with quarantines starts to weaken within a month of the introduction (Figure 6). In Colombia, people are spending less and less time in their houses, while the virus continues to spread. This contrasts with the experience in Europe, especially Italy, Spain, and France, where people left their houses only after the country had tamed the virus (Figure 9).

Figure 9. Google mobility data and contagion rates – Colombia vs. Spain

Source: Own calculations with data downloaded from Max Roser, Hannah Ritchie, Esteban Ortiz-Ospina and Joe Hasell (2020),, and Google Community Mobility Report, both downloaded 07-19-2020

Moreover, lockdowns as a policy tool in emerging countries are less sustainable due to a number of structural factors. Compared to advanced economies, it is harder to generate income during lockdowns in LATAM because informal jobs—which mostly require human contact—are hard to perform from home (Figure 10). As shown by Alfaro, Becerra, and Eslava (2020), informal jobs and small businesses are highly vulnerable to lockdowns. Thus, the inability for households to generate income is a disincentive to abide by government-mandated stay-at-home policies. In LATAM, fatigue and economic conditions make lockdowns a less effective policy option.

Figure 10. Informality and jobs that can be done from home

a) Informality vs GDP

b) Share of jobs that could be done from home

Source: panel a) Taken from Loayza (2020), panel b) taken from Dingel and Neiman (2020)

Did Colombia run out of response options too early?

Colombia’s handling of COVID-19 can be considered as a success. The first line of defense slowed down the pace of contagion and avoided excess deaths, providing additional time to strengthen the health system and increase ICU capacity. There is evidence of a flattening of the curve, but unlike countries in Europe and in the region, Colombia has not been able to bend it in the first 100 days.

The virus is still expanding. This continuous upward trend is more worrisome considering that mandatory lockdowns are now less effective. Testing is not increasing as fast as COVID is spreading, contact tracing is not working, and ICU beds are reaching high levels of occupancy rates. Given this context, local governors have introduced a stricter lockdown but enforcement is increasingly difficult (due to income-generation needs and fatigue). This situation begs the question whether Colombia ran out of bullets too soon. That is, did the country go into lockdown mode too early relative to the pandemic? Even if this were true, there are still policy choices available and some positive signs.

But first, let’s take a deeper look…

Aggregate numbers tend to hide what is happening at a more granular level. According to the Ministry of Health, close to 65 percent of municipalities have had no COVID-19 cases ever, or no positive cases in the last four weeks. Thus, the pandemic is not geographically widespread. High-impact municipalities are clustered on the Caribbean and Pacific coasts, and in the south along the border with Ecuador (Figure 11). Policies should be targeted to regions affected, avoiding unnecessary harm to local economic conditions in regions where health risks are not significant.

Figure 11. COVID-19 impact classification by municipalities

No COVID: no positive cases, Low impact: no positive cases in last 4 weeks, Moderate: positive cases in last 4 weeks, risk of high mortality or incidence, High: positive cases in last 4 cases and high incidence and mortality
Source: Ministry of Health, 07-13-2020

Also, the virus is spreading faster in larger municipalities than in rural areas (Figure 12). It is worth mentioning that Leticia, the capital of Amazonas, has reduced its rate of contagion by 87 percent , from 1,600 to 200 daily cases per million.

Figure 12. Daily cases per million inhabitants, capital cities (July 18 vs May 30)

Source: Own calculations, INS 07-18-2020

Barranquilla is now Colombia’s COVID-19 epicenter (see Figure 13). Daily cases have remained between 300 and 400 per million during the last 30 days while daily deaths peaked just below 30 per million. These numbers are staggering: Italy, at its peak, registered about 200 daily cases and 14 daily deaths per million people. Bogotá is another concern: by mid-June, daily cases have increased at a higher rate, which could mean more daily deaths in the next weeks.

Figure 13. Daily cases and deaths per million inhabitantsm, selected cities

Due to lag in the registry of deaths and cases, both graphs present data up to 14 days before the report date
Source: Own calculations with data from INS, 07-18-2020

At the start of the pandemic, Medellin and Bucaramanga had contagion and mortality rates well below other cities. Medellin’s strategy, “Medellín Me Cuida,” was based on the use of technology to gather information to isolate and support the population at risk. However, by mid-June, daily cases in Medellin began to rise much faster.

The spike in Medellin’s daily cases, per million inhabitants, coincides with the start of a consistent drop in the percentage of total cases identified through contact tracing (Figure 14). Medellin’s levels of contact tracing were consistently around 60 percent up until mid-June, when they started falling.

Figure 14. Total cases identified by contact tracing, selected towns and cities (percent)

Due to lag in the registry, the graph presents data up to 7 days before the report date
Source: Own calculations with data from INS 07-18-2019

Currently, they remain above other cities but have dropped down to 25 percent. Bucaramanga’s contact tracing also fell, before it happened in Medellin, in fact, but it is yet to translate to a higher contagion rate. Bogota and Cali’s contact tracing measures have been falling consistently since the onset of the pandemic.

Figure 15. Super contagion episodes – Villavicencio, Ituango, and El Espinal

Active cases: Total Cases – recovered – deaths
Y axis in logarithmic scale
Source: Own calculations, INS 07-18-2020

We end this section by documenting three “super contagion episodes.” These events occur when, in a short amount of time, a defined community or group of people experience an exponential increase in contagion levels. Take, for example, Korea, where it is believed that patient 31, a 60-year-old woman, when attending a religious ceremony in Daegu spread the virus through the congregation. By the end of March, 5,000 cases—more than half of Korea’s total cases at that time—were accounted for by the cluster of the Shincheonji Church of Jesus.[10]

Three super contagion episodes can be identified in Colombia. The first one happened at the beginning of March in the municipal jail of Villavicencio, where 927 people tested positive for COVID-19. The second episode occurred in the small town of El Espinal (Tolima), where, once again, COVID managed to enter the walls of the municipal jail. There, 687 positive cases were detected, most of them prisoners but also a couple dozen of guards. Moreover, the virus has managed to spread across the small town (204 cases). Positive COVID cases in El Espinal increased at a 10 percent daily rate. The last super contagion episode occurred at the hydroelectric project Hidroituango, with 760 positive cases. It is believed that contagion started following a staff change at the end of April. Fortunately, the virus has been confined to the plant, and it has not affected the areas surrounding the construction site.

These three episodes underscore that clustered outbreaks can easily occur under close-contact encounters in closed and crowded spaces (the Three Cs).[11] Fortunately, these episodes happened under confined settings, whether it was a municipal jail or a hydroelectric project, that allowed for an easier control and isolation of positive cases. However, it is plausible that these super contagion episodes might happen in public gatherings where less control is possible, as in Korea. Colombia’s government must learn from these experiences and not implement policies that drive people to Three Cs scenarios, such as the VAT-free day that was introduced by the government in June and July to stimulate retail sales.[12]

Economic policy responses

Colombia’s government and the central bank responded with an innovative policy mix. The central bank of Colombia, Banco de la República, has cut its policy rate by 175 basis points since March while providing ample liquidity both in domestic and foreign currency. Some of the tools have been to increase terms and quotas in repo operations, allowing more agents to be part of these repo operations, and by doing non-delivery forwards and foreign exchange swap auctions in US dollars. Unlike direct intervention in the foreign exchange market, these options do not imply a reduction in the stock of foreign reserves.

The central bank implemented a bond purchase program for the first time (i.e., quantitative easing). In March 23, the central bank announced that it would outright buy government and private bonds from the market. The size of this program was 0.8 percent of GDP and its goal was to inject liquidity permanently to ensure the proper operation of financial markets. According to a preliminary study by the BIS (Arslan, Drehmann, and Hofmann, 2020), the bond purchase program implemented in Colombia, and in other emerging economies, was successful in restoring investor confidence without pushing inflation or the exchange rate out of control.

Additionally, the central bank provided a COP 10 trillion (US$ 2.7 billion) relief to banks by reducing reserve requirements from 11 percent to 8 percent. The government took advantage of the extra liquidity generated by this move by forcing banks to buy the newly issued Solidarity Bonds for an amount equivalent to 1 percent of GDP.

On the fiscal front, the national government has focused its efforts on strengthening the health system (0.8 percent of GDP), providing monetary transfers to vulnerable households (0.9 percent of GDP), and minimizing the pandemic’s impact on the labor market by implementing measures to protect existing formal jobs (0.8 percent of GDP).

The government used existing cash transfer programs to provide additional support to low-income households. This infrastructure covered 3.9 million households prior to the pandemic though various social programs: Familias en Acción, Jóvenes en Acción, and Colombia Mayor. Familias en Acción households are receiving an additional COP 145K (US$ 40) in addition to their ordinary monthly transfer. This means that, on average, each beneficiary household is receiving COP 334K (US$ 90) per transfer. Jóvenes en Acción beneficiaries (tertiary education students) are getting a COP 356K (US$ 96) transfer, in addition to their ordinary transfer of COP 320K (US$ 87). And, finally, Colombia Mayor beneficiaries (population over 65) received twice the regular transfer of COP 80K (US$ 22).

On top of the existing social safety net, a new program, Ingreso Solidario, was rapidly introduced to provide unconditional cash transfers to vulnerable households that, before COVID-19, were not beneficiaries of any of the existing cash transfer programs. This program was designed to help 2.9 million poor and vulnerable households, of which 2.6 million have already been reached and are receiving COP 160K (US$ 44) per transfer. This program also fostered financial inclusion as it was necessary to open 1 million accounts, mostly through the use of digital wallets and other novel mobile phone payment services. Ingreso Solidario has been extended until the end of the year, and it is likely that this program will become part of Colombia’s permanent social safety net.

To help alleviate small and medium businesses liquidity needs, the government stepped up credit guarantees for loans aimed at supporting payrolls (90 percent of the loan is guaranteed by the government), working capital (80 percent), and independent workers (80 percent). Up to July 16, 110,000 disbursements for a total amount of COP 5 trillion (US$ 1.3 billion) have been made under this program.

Likewise, additional measures were taken by the government to help existing debtors. Any loan that wasn’t delinquent for more than 30 days by February 29 could benefit from grace periods whose length would be specific to each debtor. In turn, banks were allowed to deplete their anticyclical buffers and provisions. More than 16 million loans, belonging to 11 million different clients, and with a total value of COP 220 trillion (US$ 60 billion) have been approved by banks under this program since March.

Moreover, in an effort to prevent formal employment destruction, the government has implemented the Programa de Apoyo al Empleo Formal. This program helps formal firms with the condition that no payroll cuts can occur. Through this program, the government subsidized 40 percent of a minimum wage to each worker of any firm that has seen a 20 percent reduction in sales with respect to April of last year. This program has benefited 100,000 firms and protected 2.4 million formal jobs.

Total emergency spending amounts to 2.7 percent of GDP this year. From an initial fiscal deficit of 2.2 percent of GDP expected for 2020, and a tax revenue loss of about 3.3 percent of GDP, this extra spending will increase the expected deficit this year to 8.2 percent of GDP. To address the COVID-19 shock, the government requested that the fiscal committee suspend the fiscal rule that strictly limited deficit spending for 2020 and 2021. The committee unanimously decided to do so in June.

Figure 16. Fiscal balance (percent of GDP), central government 2020-2030 target

Gray area represents Fiscal Rule suspension period
Source: Medium Term Fiscal Framework, Ministry of Finance

In late June the Ministry of Finance presented its medium-term fiscal framework for the next 10 years. The expected deficit of 5.1 percent of GDP for 2021 is supposed to be met by cutting all the additional expenditure due to the pandemic, combined with an increase in tax revenue of 0.2 GDP percentage points and the proceeds of privatizations amounting to 1.1 percent of GDP. In addition, fiscal reform worth 2 percent of GDP is projected for 2022.

These are overly optimistic assumptions for at least four reasons:

First, dismantling in 2021 the stimulus programs implemented in 2020 will be difficult. Not only will the economy not be fully recovered, but Congress will be reluctant to approve budgetary cuts in a pre-electoral year.

Second, increases in tax collections in 2021 will be difficult, especially in the case of income taxes, given the low levels of income and corporate profits expected for 2020.

Third, privatizations are difficult and lengthy political and legal processes. The sale of 57.6 percent of the government’s stake in ISAGEN (a hydropower provider) in 2016 took six years and it was worth 0.7 percent of GDP. Selling state-owned enterprises for twice this amount in less than a year is not realistic, unless the central government is planning to sell the assets to other public sector entities, such as Ecopetrol or Financiera de Desarrollo Nacional (FDN).

Finally, a fiscal reform that increases revenue by 2 percent of GDP in 2022 is unlikely, not only because it will need to be enacted in a pre-electoral year but also because, historically, tax reforms have only been able to increase revenue by less than 1 percent of GDP.

Together, these four assumptions are unrealistic. We consider that it is the appropriate time to think about a fiscal path that is viable and credible—one that is in line with the realities of today.

First impacts and first signs of recovery

The Colombian labor market suffered a severe impact in both April and May when unemployment rates rose to 19.8 percent and 21.4 percent, respectively. The number of employed people fell by 5.5 million in April and by 4.9 million in May (relative to the same months in 2019). The number of jobs has dropped to the level observed by the end of 2006 (Figure 17).

Figure 17. Labor market indicators

Data up to May 2020
Source: Own Calculations, DANE - GEIH

Additionally, the negative impact has been more severe for women. The number of employed women fell 22.1 percent in May (year-on-year), compared to 14.9 percent for men. Women tend to have a larger participation in sector, such as entertainment or dining that are more vulnerable to lockdown measures.

The government decided to gradually reopen the economy by allowing some nonessential sectors to go back to work, starting on April 28 with construction and manufacturing, and then wholesale retail in May 11. The president has insisted that Colombians should gradually regain their productive life but not their social life. The effects of this gradual re-opening can be observed in leading indicators. For example, daily fuel demand plummeted from 6.1 million gallons of gasoline at the beginning of the year to 2.1 million gallons during the most-strict lockdown period (April). Since the gradual re-opening started, fuel demand has recovered by 105 percent, reaching 4.3 million gallons a day (see figure 18). Activity seems to be recovering but not to pre-COVID levels. Given that the pandemic situation is worse than before, it will be harder for economic activity to continue picking up—not only because local governments may reimpose lockdowns, but also because people will increase their precautionary behavior and reduce consumption.

Figure 18. National fuel demand (Daily)

Source: Reporte Semanal Ministerio de Minas y Energía Demanda Energéticos, June 19, 2020

The cost of failing is a 10-year reverse in development

We are currently facing the largest shock since the Great Depression. While during the 1999 economic crisis, Colombian GDP fell 4.4 percent, economic activity is expected to contract by 7.8 percent this year. This means going back a decade in poverty and inequality.

Two different studies, one from Jairo Núñez, from Fedesarrollo, and the other from the Department of Economics of Universidad de los Andes,[13] estimate an increase in poverty rates of 6.5 percentage points and 15 percentage points, respectively. In terms of inequality, the effect will be small for Núñez while the study by Universidad de los Andes expects that the Gini coefficient increases from 0.50 to 0.57 in the worst-case scenario. However, great uncertainty remains over these estimates since neither study considers the negative amplification effects that a prolonged recession could have on poverty rates and economic activity. Nevertheless, these estimates do show the size of the likely reversal.


Colombia managed to flatten the COVID-19 contagion rate and contain daily deaths. However, COVID-19 is yet to reach its peak. Policies which have worked before may no longer be effective. More now than ever, health and economic policies have to work hand in hand if we are to see light at the end of the tunnel. To that end, we offer six recommendations for Colombia’s response to the COVID-19 pandemiic moving forward:

  1. Promote social distancing and self-care among the population. With ICU capacity in red alert, it is crucial to promote campaigns to avoid the Three Cs: close-contact encounters in closed and crowded spaces.

  2. Continue strengthening the health care system. In practice, this means providing all the necessary personal protective equipment (PPE) in a timely manner and continuously testing health workers for COVID-19. In addition, it is necessary to speed up efforts to increase ICU capacity.

  3. Increase testing and contact tracing. Despite efforts to increase testing levels and implement a wide contact tracing policy, more needs to be done to reduce the time lag between symptoms and diagnosis, while improving enforcement in contact tracing and reduce positivity rates. With this in mind, Colombia would benefit from more quickly and widely implementing the new testing and tracing strategy (PRASS - Pruebas, Rastreo y Aislamiento Selectivo Sostenible).

  4. Differential policies. Most municipalities are COVID-free. Thus, policies should be implemented according to the need of each region. The government must have the flexibility and, with the help of local authorities, determine an adequate strategy for each context.

  5. Set up and announce a reopening plan contingent on the pandemic status. Given that lockdowns are increasingly less effective, citizens need to be informed of the set of possible paths onwards. From the start of the pandemic, national and local authorities have failed to provide a course of action beyond a week or two. In Spain, when the government decided to reopen, it established a well-defined four-phase plan where the transition between phases depended on pandemic and health indicators at the regional level.[14] This plan mapped out all possible paths ahead for more than three months. New York did the same. In Colombia, there is a lack of information about the different possible paths and the reasons behind policy choices. With more information, individuals make better decisions, and if they understand why difficult policies are necessary, they can get behind them, even lockdowns. It is time for the government to reduce policy uncertainty.

  6. Extend economic support programs in time and in beneficiaries. Ingreso Solidario excluded 1.9 million households that just missed the eligibility thresholds. These are poor and vulnerable households that need some form of government support. COVID-19 has provided evidence of the incompleteness of the social safety net. Ingreso Solidario should be made permanent and all cash transfers programs should be grouped under one program with two tiers: An unconditional basic income program for all poor and vulnerable households, with an additional transfer conditional on the number of children attending school and regular medical check-ups, and on the number of individuals over 65 years old living in the household.

    Programa de Apoyo al Empleo Formal should be extended, at least, for three more months, and its scope should expand to informal workers, in order to incentivize their transition to formality. A good way to do this is by exempting non-wage labor costs payments. The COVID-19 crisis has also shown how wage rigidity results in greater unemployment. It is important to prevent furloughs by allowing for temporary reductions in workers’ compensations.


Alfaro, Laura, Becerra, Oscar, and Eslava, Marcela (2020). EMEs and COVID-19 Shutting Down in a World of Informal and Tiny Firms. Documento CEDE, (19).

Arslan, Y., Drehmann, M., and Hofmann, B. (2020). Central bank bond purchases in emerging market economies (No. 20). Bank for International Settlements.

COVID-19 Community Mobility Reports – Google

Dingel, J. I., and Neiman, B. (2020). How many jobs can be done at home? (No. w26948). National Bureau of Economic Research.

Financial Times. Coronavirus tracker page

Hale, Thomas, Sam Webster, Anna Petherick, Toby Phillips, and Beatriz Kira (2020). Oxford COVID-19 Government Response Tracker, Blavatnik School of Government. Data use policy: Creative Commons Attribution CC BY standard.

Loayza, Norman. V. (2020). Costs and Trade-Offs in the Fight Against the COVID-19 Pandemic: A Developing Country Perspective.

Nota Macroeconómica No. 20 (2020) Efectos en pobreza y desigualdad del Covid-19 en Colombia: un retroceso de dos décadas. Observatorio de coyuntura económica y social del Centro de Estudios sobre Desarrollo Económico – CEDE de la Facultad de Economía de la Universidad de los Andes

Núñez, Jairo (2020) Efecto del Aislamiento por Covid-19. Work in progress

Roser, Mark, Hannah Ritchie, Esteban Ortiz-Ospina and Joe Hasell (2020) - "Coronavirus Pandemic (COVID-19)". Published online at Retrieved from: '' [Online Resource]


[1] Published every day by Instituto Nacional de Salud (

[2] Number of deaths over number of confirmed cases

[3] The Financial Times defines excess mortality as “the numbers of deaths over and above the historical average.”

[11] The government of Japan was the first to push forward a campaign for people to avoid/minimize settings with the Three Cs. See

[13] The former takes into account government policies and cash transfers while the latter does not.

Rights & Permissions

You may use and disseminate CGD’s publications under these conditions.